Repository navigation
Conversation
Port the `plan` and `review` skills from graphistrygpt into `.agents/skills/` and retarget them to this repo: - plan: repo validation commands (flake8 lint.sh + pytest, docker build) - review: drop frontend/TS, authz-parity, k8s/health-probe and notebook specifics; add CI/release supply-chain (GitHub Actions) and CloudFormation deployment-template dimensions, Streamlit view checks, repo placement rules - anti_pattern_scan.sh: new signal catalog for Python, env files, workflows (unpinned actions, missing permissions, script injection, long-lived AWS keys, risky triggers, ungated secrets) and templates; default range origin/master...HEAD; fix pipefail/SIGPIPE bugs that misreported added lines as pre-existing on large diffs and aborted the scan on `rg | head` - AGENTS.md pointer, gitignore plans/, changelog Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
…is empty Under set -e, `[ -n "" ] && ...` as the last command of the eval'd loop returned 1 and silently exited the scan for any range without e.g. Python or env files. Found while running the review skill on the stacked publish PR. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
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
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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
Adds repo-local agent skills under
.agents/skills/, ported from graphistrygpt and adapted for graph-app-kit:plan: opt-in, file-based task plans (plans/<task>/plan.md) with anti-drift rules for multi-session / multi-PR work. Validation commands now point at this repo's tooling (src/python/bin/lint.sh+ pytest, docker build).review: multi-wave PR review. Each dimension and file gets its own findings, every finding gets an adversarial verification pass, and the loop runs until findings converge. Adapted by:permissions:, OIDC vs long-lived keys, environment gates, SHA pinning, script injection. This matters becausepublish.ymlanddockerhubpublish.ymlship customer-facing artifacts.src/bootstraps/src/envs/*.env→ compose →os.environ)anti_pattern_scan.sh: a new signal catalog (AP003/009/010/011 for Python, AP016 for committed env creds, AP020–AP026 for workflows, AP030 for templates). Default range isorigin/master...HEAD. Also fixes two pipefail/SIGPIPE bugs in the upstream script:printf | grep -qmarked every line[pre-existing]on large diffsrg | head -1inside$()aborted the scan underset -eAGENTS.mdpointer;plans/gitignored; CHANGELOG entry.A stacked follow-up PR uses these skills to fix the P0 publish workflow exposure.
Verification
bash .agents/skills/review/tests/test_verdict_triage.sh→[ok] verdict triage tests passedbash .agents/skills/review/scripts/lint_anti_pattern_suppressions.sh→[ok]bash .agents/skills/review/scripts/anti_pattern_scan.sh 4b825dc642cb6eb9a060e54bf8d69288fbee4904(whole repo as added): 80 signals. Every expected hit on the currentpublish.ymlfires: AP020checkout@v3/configure-aws-credentials@v1, AP023 stored AWS keys, AP024labeledtrigger with secrets, AP025 dispatch with no environment, AP026 secrets inrun:. AP030 flags the documented SG ingress ports. No crash.Docs/tooling only. No runtime or workflow behavior changes.
🤖 Generated with Claude Code