Improve missing-env workflow recovery#49
Merged
Conversation
Contributor
📝 WalkthroughWalkthroughImproved detection/extraction of missing-environment messages, allow MISSING_ENV_VAR into the auto-fix path, add a deterministic preflight that injects a repo-local .env loader and assertions into generated workflows, update persona repair requirements and generation, and extend tests and docs. ChangesDeterministic missing-env detection → auto-repair → generated workflow loader
Sequence Diagram(s)sequenceDiagram
actor Runtime
participant Entrypoint
participant Repairer
participant Generator
participant WorkflowArtifact
participant Filesystem
Runtime->>Entrypoint: emit stderr / evidence
Entrypoint->>Entrypoint: hasMissingEnvironmentMessage + extractMissingEnvVars
Entrypoint-->>Repairer: report MISSING_ENV_VAR evidence
Repairer->>Generator: repairWorkflowDeterministically(evidence)
Generator->>WorkflowArtifact: inject loadRickyWorkflowEnv() & assertions
WorkflowArtifact->>Filesystem: read .env.local / .env at runtime (loader)
Repairer->>Runtime: retry startFromStep: runtime-launch with repaired artifact
Runtime->>Entrypoint: run proceeds / success or new evidence
Estimated code review effort🎯 4 (Complex) | ⏱️ ~45 minutes Possibly related PRs
Poem
🚥 Pre-merge checks | ✅ 4 | ❌ 1❌ Failed checks (1 warning)
✅ Passed checks (4 passed)
✏️ Tip: You can configure your own custom pre-merge checks in the settings. ✨ Finishing Touches📝 Generate docstrings
🧪 Generate unit tests (beta)
Comment |
![chatgpt-codex-connector[bot]](https://avatars.githubusercontent.com/in/1144995?s=60&v=4){kind=small}
There was a problem hiding this comment.
💡 Codex Review
Here are some automated review suggestions for this pull request.
Reviewed commit: 83d63acd69
ℹ️ About Codex in GitHub
Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you
- Open a pull request for review
- Mark a draft as ready
- Comment "@codex review".
If Codex has suggestions, it will comment; otherwise it will react with 👍.
Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".
Comment on lines
+2098
to
+2100
| const context = line.slice(Math.max(0, index - 100), index + name.length + 100).toLowerCase(); | ||
| const hasEnvContext = /env(?:ironment)?(?:\s+variable)?|variable/.test(context); | ||
| const hasMissingContext = /missing|required|not\s+set|unset/.test(context); |
There was a problem hiding this comment.
When an explicit missing-env line also mentions any uppercase workflow/product term within 100 characters, this context window makes that term look like a missing variable too. For example, missing env var GITHUB_TOKEN before calling API produces recovery steps for both GITHUB_TOKEN and API, so the false export API=... guidance this change is meant to prevent can still occur whenever the unrelated term is on the same line as the real missing-env message.
Useful? React with 👍 / 👎.
![coderabbitai[bot]](https://avatars.githubusercontent.com/in/347564?s=60&v=4){kind=small}
Contributor
There was a problem hiding this comment.
Actionable comments posted: 1
🤖 Prompt for all review comments with AI agents
Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.
Inline comments:
In `@src/local/entrypoint.ts`:
- Around line 2074-2109: hasMissingEnvironmentMessage and extractMissingEnvVars
are too permissive and allow unrelated ALL-CAPS tokens (like API) to be treated
as env vars; tighten the parsing so only names syntactically tied to an
env-message are returned. Change hasMissingEnvironmentMessage to require an
explicit env/variable token (env, environment, variable) when matching the "NAME
is required/missing/not set" pattern rather than treating any ALL-CAPS + "is
required" line as a missing-env line. In extractMissingEnvVars, when iterating
matches in a line, restrict the env-context check to a small local window around
the match (e.g., ±20 chars) and require an env/variable keyword inside that
window (instead of scanning the whole line), and only add the name if that local
context also contains a missing-state token (missing|required|not set|unset);
keep excluding known non-env tokens like ENOENT/PATH. Apply these changes inside
the functions hasMissingEnvironmentMessage and extractMissingEnvVars.
🪄 Autofix (Beta)
Fix all unresolved CodeRabbit comments on this PR:
- Push a commit to this branch (recommended)
- Create a new PR with the fixes
ℹ️ Review info
⚙️ Run configuration
Configuration used: Organization UI
Review profile: CHILL
Plan: Pro Plus
Run ID: 1b19a4fd-700e-4c8a-976c-7c6d3bd913ee
📒 Files selected for processing (2)
src/local/entrypoint.test.tssrc/local/entrypoint.ts
khaliqgant
changed the title
Contributor
There was a problem hiding this comment.
Actionable comments posted: 2
♻️ Duplicate comments (1)
src/local/entrypoint.ts (1)
2074-2115:⚠️ Potential issue | 🟠 Major | ⚡ Quick winTighten this parser to explicit missing-env syntax only.
The current heuristic still leaves the original false-positive path open. For example,
MISSING_ENV_VAR: TEST_TOKENwill add the sentinel itself because Line 2093 capturesMISSING_ENV_VARand Line 2099 treats theENVsubstring as env context, so the blocker can still suggestexport MISSING_ENV_VAR=.... Line 2079 also still classifies uppercase prose likeAPI is required, and Line 2076 does not skip negated forms likeno missing env variables. Please switch both helpers to parse only names that are syntactically attached to an explicit env/variable phrase instead of scanning whole lines for ALL_CAPS tokens.🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@src/local/entrypoint.ts` around lines 2074 - 2115, hasMissingEnvironmentMessage and extractMissingEnvVars are too permissive — update both to only detect explicit "env/variable" syntax (and skip negations) instead of scanning whole lines for ALL_CAPS tokens; change hasMissingEnvironmentMessage to reject lines with negations like "not missing" or "no missing" and only return true for tight patterns such as "(env|environment|variable|env var|process.env\\.|\\$)\\s*(NAME|\\bNAME=\\b|NAME\\b).* (missing|required|not set|unset)" or "NAME\\s*(=|:)\\s*.*" where NAME is the identifier, and change extractMissingEnvVars to stop global ALL_CAPS scanning and only add a name when it is matched by those explicit adjacent patterns (or explicit forms process.env.NAME, $NAME, NAME=, or "NAME is missing"), referencing functions extractMissingEnvVars, hasMissingEnvironmentMessage and looksLikeConfigEnvName so you locate and replace the broad regex logic with these stricter, proximity-based regex checks.
🤖 Prompt for all review comments with AI agents
Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.
Inline comments:
In `@src/local/auto-fix-loop.test.ts`:
- Around line 404-408: The retry metadata currently propagates synthetic
launch-time step IDs (e.g., 'runtime-launch', 'runtime-precheck') from
auto-fix-loop.ts → entrypoint.ts → local-coordinator.ts into the retry object
used by runSingleAttempt/agent-relay; filter these out when building retry
metadata in the place where retry is constructed (the code that creates the
retry object consumed by runSingleAttempt) so that for launch-time blockers
(MISSING_ENV_VAR, MISSING_BINARY) you either omit startFromStep entirely or drop
any startFromStep that matches the synthetic IDs
('runtime-launch'|'runtime-precheck') before passing to
local-coordinator/agent-relay. Ensure the change is applied where retry is
assembled (referencing the retry object passed into runSingleAttempt) so tests
expecting no synthetic startFromStep (or omitted startFromStep) will pass.
In `@src/product/generation/template-renderer.ts`:
- Around line 831-855: unquoteRickyWorkflowEnvValue currently only trims and
unquotes values so unquoted inline comments like `FOO=bar # local` become part
of the value; update unquoteRickyWorkflowEnvValue to, after trimming and
determining the value is unquoted, strip any inline comment by removing the
first `#` and everything after it when the `#` is preceded by whitespace (or at
line start), then trim the result before returning; keep the existing branch
that returns the exact inner content for quoted values so `#` inside quotes is
preserved. Ensure you only change unquoteRickyWorkflowEnvValue (referenced by
loadRickyWorkflowEnv) so the loader will normalize unquoted `.env` values
correctly.
---
Duplicate comments:
In `@src/local/entrypoint.ts`:
- Around line 2074-2115: hasMissingEnvironmentMessage and extractMissingEnvVars
are too permissive — update both to only detect explicit "env/variable" syntax
(and skip negations) instead of scanning whole lines for ALL_CAPS tokens; change
hasMissingEnvironmentMessage to reject lines with negations like "not missing"
or "no missing" and only return true for tight patterns such as
"(env|environment|variable|env
var|process.env\\.|\\$)\\s*(NAME|\\bNAME=\\b|NAME\\b).* (missing|required|not
set|unset)" or "NAME\\s*(=|:)\\s*.*" where NAME is the identifier, and change
extractMissingEnvVars to stop global ALL_CAPS scanning and only add a name when
it is matched by those explicit adjacent patterns (or explicit forms
process.env.NAME, $NAME, NAME=, or "NAME is missing"), referencing functions
extractMissingEnvVars, hasMissingEnvironmentMessage and looksLikeConfigEnvName
so you locate and replace the broad regex logic with these stricter,
proximity-based regex checks.
🪄 Autofix (Beta)
Fix all unresolved CodeRabbit comments on this PR:
- Push a commit to this branch (recommended)
- Create a new PR with the fixes
ℹ️ Review info
⚙️ Run configuration
Configuration used: Organization UI
Review profile: CHILL
Plan: Pro Plus
Run ID: c4347ba8-0e4c-4b18-9bdd-a3b01d6d0887
📒 Files selected for processing (9)
src/local/auto-fix-loop.test.tssrc/local/auto-fix-loop.tssrc/local/entrypoint.test.tssrc/local/entrypoint.tssrc/product/generation/pipeline.test.tssrc/product/generation/template-renderer.tssrc/product/generation/workforce-persona-repairer.tssrc/product/generation/workforce-persona-writer.tsworkflows/shared/WORKFLOW_AUTHORING_RULES.md
✅ Files skipped from review due to trivial changes (1)
- workflows/shared/WORKFLOW_AUTHORING_RULES.md
🚧 Files skipped from review as they are similar to previous changes (1)
- src/local/entrypoint.test.ts
Comment on lines
+404
to
+408
| expect(runSingleAttempt.mock.calls[1][0].retry).toMatchObject({ | ||
| previousRunId: 'run-1', | ||
| retryOfRunId: 'run-1', | ||
| startFromStep: 'runtime-launch', | ||
| }); |
Contributor
There was a problem hiding this comment.
🧩 Analysis chain
🏁 Script executed:
#!/bin/bash
set -euo pipefail
echo "Inspect retry construction in src/local/auto-fix-loop.ts"
sed -n '1,260p' src/local/auto-fix-loop.ts | nl -ba
echo
echo "Inspect synthetic failed-step producers/consumers"
rg -n -C3 '\bruntime-launch\b|\bstartFromStep\b|\bfailed_step\b' src/local/entrypoint.ts src/local/auto-fix-loop.tsRepository: AgentWorkforce/ricky
Length of output: 182
🏁 Script executed:
#!/bin/bash
set -euo pipefail
echo "=== Inspect auto-fix-loop.ts: retry construction (lines 1-260) ==="
head -260 src/local/auto-fix-loop.ts
echo ""
echo "=== Search for runtime-launch and startFromStep usage ==="
rg -n -C3 'runtime-launch|startFromStep|failed_step' src/local/entrypoint.ts src/local/auto-fix-loop.tsRepository: AgentWorkforce/ricky
Length of output: 17833
🏁 Script executed:
#!/bin/bash
set -euo pipefail
echo "=== Search for failedStepFromEvidence function definition ==="
rg -n 'failedStepFromEvidence' src/
echo ""
echo "=== Search for any remapping/filtering of runtime-launch ==="
rg -n 'runtime-launch|runtime-precheck' src/ -B2 -A2
echo ""
echo "=== Check test file context around lines 404-408 ==="
sed -n '390,420p' src/local/auto-fix-loop.test.tsRepository: AgentWorkforce/ricky
Length of output: 7960
🏁 Script executed:
#!/bin/bash
set -euo pipefail
echo "=== Check failedStepFromEvidence implementation (lines 1348-1370) ==="
sed -n '1348,1370p' src/local/auto-fix-loop.ts
echo ""
echo "=== Check if coordinator/runner has handling for synthetic step IDs ==="
fd -t f 'coordinator' src/ | head -5
echo ""
echo "=== Search for any filtering logic before passing to coordinator ==="
rg -n 'coordinator.launch|--start-from' src/local/entrypoint.ts -B3 -A3Repository: AgentWorkforce/ricky
Length of output: 2137
🏁 Script executed:
#!/bin/bash
set -euo pipefail
echo "=== Search local-coordinator.ts for startFrom handling ==="
rg -n 'startFrom|start.from|start-from' src/runtime/local-coordinator.ts -B2 -A2
echo ""
echo "=== Check if there's any remapping in coordinator ==="
cat src/runtime/local-coordinator.ts | head -100Repository: AgentWorkforce/ricky
Length of output: 3737
🏁 Script executed:
#!/bin/bash
set -euo pipefail
echo "=== Check if there's filtering for synthetic step IDs anywhere ==="
rg -n 'runtime-launch|runtime-precheck' src/runtime/ -B2 -A2
echo ""
echo "=== Check agent-relay documentation or handling of --start-from ==="
fd -t f 'agent-relay' src/ 2>/dev/null || echo "agent-relay not found in src"
echo ""
echo "=== Check if startFromStep is conditionally set based on step ID ==="
rg -n 'startFromStep.*=|startFromStep\?' src/local/ -B2 -A2Repository: AgentWorkforce/ricky
Length of output: 131
🏁 Script executed:
#!/bin/bash
set -euo pipefail
echo "=== Look at the test helper blockerResponse to understand how runtime-launch gets set ==="
rg -n 'blockerResponse.*runtime-launch|function blockerResponse' src/local/auto-fix-loop.test.ts -B2 -A15
echo ""
echo "=== Check what happens in the SDK runner when --start-from is called with invalid step ==="
fd -t f 'runner|relay' src/product/ | head -3
echo ""
echo "=== Check if there's any step validation or filtering before execution ==="
rg -n 'startFromStep|start-from' src/product/ -B1 -A1Repository: AgentWorkforce/ricky
Length of output: 4796
🏁 Script executed:
#!/bin/bash
set -euo pipefail
echo "=== Look at the full blockerResponse function ==="
sed -n '580,620p' src/local/auto-fix-loop.test.ts
echo ""
echo "=== Check how the test evidence is constructed ==="
rg -n 'missingEnvEvidence|function.*Evidence' src/local/auto-fix-loop.test.ts -A20 | head -60Repository: AgentWorkforce/ricky
Length of output: 4420
🏁 Script executed:
#!/bin/bash
set -euo pipefail
echo "=== Check if agent-relay or coordinator ignores invalid startFromStep ==="
rg -n 'start.from|startFromStep|--start-from' src/runtime/local-coordinator.ts | head -20
echo ""
echo "=== Check what buildInvocationSummary does with startFromStep ==="
sed -n '275,340p' src/runtime/local-coordinator.ts
echo ""
echo "=== Look for any documentation or tests about synthetic step IDs ==="
rg -n 'runtime-launch.*resume|synthetic.*step|step.*validation' src/ -iRepository: AgentWorkforce/ricky
Length of output: 6384
🏁 Script executed:
#!/bin/bash
set -euo pipefochop pipefail
echo "=== Check if there's any agent-relay handling or documentation of synthetic steps ==="
rg -n 'agent-relay|--start-from' README.md CONTRIBUTING.md 2>/dev/null || echo "No docs found"
echo ""
echo "=== Check test to see if second call actually validates the retry was properly formed ==="
sed -n '375,420p' src/local/auto-fix-loop.test.ts
echo ""
echo "=== Verify: does the second mockResolvedValue actually test agent-relay behavior or just return success? ==="
rg -n 'successResponse|mockResolvedValue' src/local/auto-fix-loop.test.ts | grep -A2 -B2 'successResponse.*run-2'Repository: AgentWorkforce/ricky
Length of output: 120
🏁 Script executed:
#!/bin/bash
set -euo pipefail
echo "=== Check if there's any agent-relay handling or documentation of synthetic steps ==="
rg -n 'agent-relay|--start-from' README.md CONTRIBUTING.md 2>/dev/null || echo "No docs found"
echo ""
echo "=== Check test to see if second call actually validates the retry was properly formed ==="
sed -n '375,420p' src/local/auto-fix-loop.test.ts
echo ""
echo "=== Verify: does the second mockResolvedValue actually test agent-relay behavior or just return success? ==="
rg -n 'successResponse|mockResolvedValue' src/local/auto-fix-loop.test.ts | grep -A2 -B2 'successResponse.*run-2'Repository: AgentWorkforce/ricky
Length of output: 4865
🏁 Script executed:
#!/bin/bash
set -euo pipefail
echo "=== Final check: Is there any remapping between failedStep extraction and retry construction? ==="
sed -n '115,230p' src/local/auto-fix-loop.ts | grep -A10 -B10 'startFromStep'
echo ""
echo "=== Confirm: agent-relay does NOT have special handling for runtime-launch ==="
fd -t f 'relay' src/ 2>/dev/null | head -5 || echo "agent-relay source not in repo"
echo ""
echo "=== Check if there's any comment or documentation about resuming from synthetic steps ==="
rg -n 'runtime-launch.*resume|resume.*synthetic|step.*synthetic' src/ -B2 -A2Repository: AgentWorkforce/ricky
Length of output: 1011
Filter out synthetic step IDs before constructing retry metadata.
The test now expects startFromStep: 'runtime-launch', but runtime-launch is a synthetic label from src/local/entrypoint.ts used only to mark launch-time blockers—it is not an actual workflow step. Currently, this synthetic ID flows unchanged through auto-fix-loop.ts → entrypoint.ts → local-coordinator.ts and reaches agent-relay as --start-from runtime-launch, where it will fail because the step does not exist in the workflow.
For launch-time blockers (MISSING_ENV_VAR, MISSING_BINARY), either:
- Omit
startFromStepfrom the retry metadata entirely, or - Filter out synthetic step IDs (e.g.,
runtime-launch,runtime-precheck) before passing to the retry constructor.
Test expectation showing the issue
expect(runSingleAttempt.mock.calls[1][0].retry).toMatchObject({
previousRunId: 'run-1',
retryOfRunId: 'run-1',
startFromStep: 'runtime-launch',
});
🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.
In `@src/local/auto-fix-loop.test.ts` around lines 404 - 408, The retry metadata
currently propagates synthetic launch-time step IDs (e.g., 'runtime-launch',
'runtime-precheck') from auto-fix-loop.ts → entrypoint.ts → local-coordinator.ts
into the retry object used by runSingleAttempt/agent-relay; filter these out
when building retry metadata in the place where retry is constructed (the code
that creates the retry object consumed by runSingleAttempt) so that for
launch-time blockers (MISSING_ENV_VAR, MISSING_BINARY) you either omit
startFromStep entirely or drop any startFromStep that matches the synthetic IDs
('runtime-launch'|'runtime-precheck') before passing to
local-coordinator/agent-relay. Ensure the change is applied where retry is
assembled (referencing the retry object passed into runSingleAttempt) so tests
expecting no synthetic startFromStep (or omitted startFromStep) will pass.
Comment on lines
+831
to
+855
| function renderWorkflowEnvLoaderHelper(): string { | ||
| return `function loadRickyWorkflowEnv(cwd = process.cwd()) { | ||
| for (const file of ['.env.local', '.env']) { | ||
| const path = rickyWorkflowPath.join(cwd, file); | ||
| if (!rickyWorkflowFs.existsSync(path)) continue; | ||
| const body = rickyWorkflowFs.readFileSync(path, 'utf8'); | ||
| for (const rawLine of body.split(/\\r?\\n/)) { | ||
| const line = rawLine.trim(); | ||
| if (!line || line.startsWith('#')) continue; | ||
| const match = /^(?:export\\s+)?([A-Za-z_][A-Za-z0-9_]*)\\s*=\\s*(.*)$/.exec(line); | ||
| if (!match) continue; | ||
| const [, key, rawValue] = match; | ||
| if (!key || rawValue === undefined || process.env[key] !== undefined) continue; | ||
| process.env[key] = unquoteRickyWorkflowEnvValue(rawValue); | ||
| } | ||
| } | ||
| } | ||
|
|
||
| function unquoteRickyWorkflowEnvValue(value: string): string { | ||
| const trimmed = value.trim(); | ||
| if ((trimmed.startsWith('"') && trimmed.endsWith('"')) || (trimmed.startsWith("'") && trimmed.endsWith("'"))) { | ||
| return trimmed.slice(1, -1); | ||
| } | ||
| return trimmed; | ||
| } |
Contributor
There was a problem hiding this comment.
Strip inline comments from unquoted .env values.
Right now FOO=bar # local is loaded as bar # local, because the helper only trims and unwraps quotes. That will corrupt common .env files and break the values this new loader is supposed to recover. Please normalize unquoted values before assigning them, while still preserving # inside quoted strings.
🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.
In `@src/product/generation/template-renderer.ts` around lines 831 - 855,
unquoteRickyWorkflowEnvValue currently only trims and unquotes values so
unquoted inline comments like `FOO=bar # local` become part of the value; update
unquoteRickyWorkflowEnvValue to, after trimming and determining the value is
unquoted, strip any inline comment by removing the first `#` and everything
after it when the `#` is preceded by whitespace (or at line start), then trim
the result before returning; keep the existing branch that returns the exact
inner content for quoted values so `#` inside quotes is preserved. Ensure you
only change unquoteRickyWorkflowEnvValue (referenced by loadRickyWorkflowEnv) so
the loader will normalize unquoted `.env` values correctly.
Contributor
There was a problem hiding this comment.
♻️ Duplicate comments (2)
src/local/entrypoint.ts (2)
2074-2080:⚠️ Potential issue | 🟠 Major | ⚡ Quick winHandle negated missing-env lines before matching positives.
Line 2077 only ignores
not missing env...; lines likeNo missing environment variablesstill match Line 2078 and can misclassify blockers asMISSING_ENV_VARvia Line 2003.Suggested fix
function hasMissingEnvironmentMessage(text: string): boolean { for (const rawLine of text.split(/\r?\n/)) { const line = stripAnsi(rawLine); - if (/\bnot\s+missing\s+env(?:ironment)?\b/i.test(line)) continue; + if (/\b(?:no|not)\s+missing\s+env(?:ironment)?(?:\s+var(?:iable)?s?)?\b/i.test(line)) continue; if (/(?:missing|required).{0,80}(?:env(?:ironment)?(?:\s+variable)?)/i.test(line)) return true; if (/(?:env(?:ironment)?(?:\s+variable)?).{0,80}(?:missing|required|not\s+set|unset)/i.test(line)) return true; } return false; }🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@src/local/entrypoint.ts` around lines 2074 - 2080, The hasMissingEnvironmentMessage function currently only skips lines matching "not missing env..." but still counts messages like "No missing environment variables" as positive; update hasMissingEnvironmentMessage to detect common negation patterns (e.g., "no missing", "none missing", "no .* missing", "no missing environment variables", "environment variables are not missing") before the positive regex checks and skip those lines (continue) so they don't trigger the positive matches; modify the checks inside hasMissingEnvironmentMessage to first run a negation-regex branch and then the existing positive/required env regexes.
2120-2123:⚠️ Potential issue | 🟠 Major | ⚡ Quick winAlign env-name acceptance with concrete-var filtering to avoid acronym false positives.
Line 2122 still admits short all-caps tokens like
API/MSD, which can leak into blockermissingandexport ...recovery steps even when they’re prose. This conflicts with stricter filtering used insrc/local/auto-fix-loop.ts(isConcreteEnvVarName, Lines 513-517 in provided context).Suggested fix
function addEnvNames(names: Set<string>, value: string | undefined): void { for (const name of (value ?? '').split(/\s*,\s*/)) { - if (/^[A-Z][A-Z0-9_]{2,}$/.test(name) && !['ENOENT', 'PATH'].includes(name)) { + if ( + /^[A-Z][A-Z0-9_]{2,}$/.test(name) && + !['ENOENT', 'PATH'].includes(name) && + (name.includes('_') || name.length > 3) + ) { names.add(name); } } }🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@src/local/entrypoint.ts` around lines 2120 - 2123, The addEnvNames function is allowing short all-caps tokens (e.g., "API") that the concrete-var filter disallows; instead of the current inline regex in addEnvNames, call the shared isConcreteEnvVarName(name) predicate (import it from its module) and use that for the acceptance check (while still excluding ['ENOENT','PATH']) so env-name acceptance matches the stricter concrete-var filtering used in auto-fix-loop.
🤖 Prompt for all review comments with AI agents
Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.
Duplicate comments:
In `@src/local/entrypoint.ts`:
- Around line 2074-2080: The hasMissingEnvironmentMessage function currently
only skips lines matching "not missing env..." but still counts messages like
"No missing environment variables" as positive; update
hasMissingEnvironmentMessage to detect common negation patterns (e.g., "no
missing", "none missing", "no .* missing", "no missing environment variables",
"environment variables are not missing") before the positive regex checks and
skip those lines (continue) so they don't trigger the positive matches; modify
the checks inside hasMissingEnvironmentMessage to first run a negation-regex
branch and then the existing positive/required env regexes.
- Around line 2120-2123: The addEnvNames function is allowing short all-caps
tokens (e.g., "API") that the concrete-var filter disallows; instead of the
current inline regex in addEnvNames, call the shared isConcreteEnvVarName(name)
predicate (import it from its module) and use that for the acceptance check
(while still excluding ['ENOENT','PATH']) so env-name acceptance matches the
stricter concrete-var filtering used in auto-fix-loop.
ℹ️ Review info
⚙️ Run configuration
Configuration used: Organization UI
Review profile: CHILL
Plan: Pro Plus
Run ID: 56b029d9-ff5d-412f-92b6-a787100f4c84
📒 Files selected for processing (2)
src/local/entrypoint.test.tssrc/local/entrypoint.ts
✅ Files skipped from review due to trivial changes (1)
- src/local/entrypoint.test.ts
This was referenced May 6, 2026
This was referenced May 14, 2026
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
Test Plan