retry-loop
/retry-loop
Use when the same error message recurs after a fix attempt, or this is the 2nd+ try at the same fix. Stop varying the fix, state the exact error and the theory, then ask instead of trying a third variation. Includes an attempt-ledger protocol.
- Author
- moah0911 ↗
- Repository
- aftermath ↗
- Licence
- MIT
- Last repository change
- 07.11.2026
Assessment and security
- Files in this package
- 1
- Scripts
- 0
- HTML comments
- 0
- URLs in SKILL.md
- 0
No signals detected. Repository changed on 07.11.2026, without references to retired models or APIs.
MIT, open: use and modify while retaining author attribution.
What it does and when to use it
Use when the same error message recurs after a fix attempt, or this is the 2nd+ try at the same fix. Stop varying the fix, state the exact error and the theory, then ask instead of trying a third variation. Includes an attempt-ledger protocol.
Original author description
Use when the same error message recurs after a fix attempt, or this is the 2nd+ try at the same fix. Stop varying the fix, state the exact error and the theory, then ask instead of trying a third variation. Includes an attempt-ledger protocol.
How it works
- Attempt ledger
- Steps
- Forbidden
SKILL.md preview
--- name: retry-loop description: > Use when the same error message recurs after a fix attempt, or this is the 2nd+ try at the same fix. Stop varying the fix, state the exact error and the theory, then ask instead of trying a third variation. Includes an attempt-ledger protocol. license: MIT --- # Retry Loop Same error twice means the theory is wrong, not the formatting. ## Attempt ledger Keep one line per attempt. The distinction that matters: - Same error, no edit in between → same attempt. Re-running did not test a new theory. - Same error after an edit → 2nd failed attempt. Stop here. - Green test, `/aftermath-revert`, or `/aftermath off` → ledger resets. Format: ``` attempt 1 | changed: [hunk/flag] | error: [exact] | proved: [what it ruled out] attempt 2 | changed: [hunk/flag] | error: [exact] | proved: [what it ruled out] ``` ## Steps 1. Stop. No third variation. 2. Paste the exact error verbatim — not a paraphrase. Trim only repeated stack frames. 3. State the current theory in one line, labeled as theory until verified. 4. State what was already tried (attempt 1, attempt 2) and what each attempt proved. 5. Ask: confirm the theory with the user, or request the missing fact (log, version, repro step) instead of guessing it. 6. One deliberate next step only after the answer. If it fails identically, return to step 2 — do not widen scope. ## Forbidden - Varying imports, flags, or syntax hoping the error changes shape. - Filling a missing fact with a guess. - Touching files outside the failing path to "see if it helps". - Re-running the same command and calling it a new attempt. - Disguising a command to get past an enforcement block. Gap-filling must be said out loud. If the second attempt fails identically, the next move is a question, not attempt three.
Versions
The database has no other skills named retry-loop.
Database snapshot: 13.11.2026 · package version 5e55f3e586a4 (SHA-256 digest of the entire package)
Downloads in the last 30 days: 0 · ranking
Report a problem with this skill: harmful content, incorrect descriptions or an author’s removal request.
Similar skills
- 100fastly-kv-concurrent-write-retry
Fix intermittent 500 errors with "Failed to store" in Fastly KV Store during concurrent writes. Use when: (1) Batch operations fail with ~10-20% error rate, (2) Error contains "Failed to store list" or similar KV write failures, (3) Multiple requests updating the same KV key… - 90unstick
The app broke and you did not write the code -- find the actual cause instead of guessing at fixes, and stop the loop where every attempted fix adds a new bug. Use when something worked yesterday, the build fails, the deploy is white, or the same error keeps coming back. - 100self-heal
Autonomous diagnose-research-fix-verify loop — up to 5 attempts before human escalation. Use when: test failing, error, self-heal, stuck, broken, can't fix, autonomous fix, retry, fix this. - 100verify-and-fix
This skill should be used when the user reports an error, bug, or unexpected behavior and wants it diagnosed and fixed. Trigger on phrases like "check this error", "check this bug", "here's an error", "here's a bug", "I have an error", "I have a bug", "found a bug", "got an… - 90root-cause-first
Debugging discipline. Use whenever diagnosing a bug, error, test failure, crash, regression, flaky behavior, or unexpected output, and especially before applying any fix. Also trigger when a previous fix attempt did not work and another attempt is about to be made.