llm-self-heal SKILL.md skillA
llm-self-heal from enricai/leerie is an agent skill: a SKILL.md folder an AI agent loads when a task calls for it.
mdr add enricai/leerie/llm-self-heal@git:20260702.2f2185fPins git:20260702.2f2185f in mdr.lock. Pin the fingerprint instead.
- Audit grade
- A, 17 of 17 checks passed
- Kind
- SKILL.md skill
- Repository
- enricai/leerie
- Path
skills/llm-self-heal/SKILL.md- Versions recorded
- 5
- Latest version
- git:20260702.2f2185f, committed Jul 2, 2026
- First version recorded
- May 28, 2026
- Size
- 12,191 bytes
- Repository stars
- 10
- License
- MIT
Autonomous self-healing loop for leerie worker prompts that produced captured failures. For each call_type with failures, runs a measured n=N baseline (unpatched), then iterates: invoke patch-generator worker (claude -p) → apply proposed patch → replay patched arm → score → check convergence (SUCCESS/PLATEAUED/BUDGET_EXHAUSTED/TIMEOUT/REGRESSED). Writes a healing-<call_type>.md report per call_type with the best patch and the iteration history. Production prompts in prompts/ stay manual — the…
Collected from public GitHub. Every version is saved exactly as it was and never changes. Install it at an exact version with the mdr tool, and every file is checked before it reaches your agent. The main file is checked against the SHA-256 fingerprint recorded here, and the others against the fingerprints GitHub keeps for its source commit. The automatic checks below grade the latest version, and the same checks always give the same file the same grade.
What the file says
<objective> Drive the autonomous heal loop for one or more leerie `call_type`s that produced failures in a judge-llm-batch run. The loop iterates: 1. **Baseline** — run n=N unpatched replays per failing sample via `claude -p`, score each, establish noise floor. 2. **Loop** — invoke the patch-generator worker (a direct `claude -p` call) to propose a minimal patch to the system prompt, apply the patch, replay the patched arm, score, check convergence. 3. **Report** — write `<heal-dir>/<call_type>/healing-<call_type>.md` with the verdict (SUCCESS / PLATEAUED / BUDGET_EXHAUSTED / TIMEOUT / REGRESSED), the best patch found, and the full iteration history. **Output:** per call_type with failures, a heal report under the run's `<heal_subdir>/` directory (default `heal-out/`; configurable via `--heal-dir` / `LEERIE_HEAL_DIR` / `leerie.toml heal_dir`). Production prompts in `prompts/` are NOT modified by this skill. Patches are proposed evidence — applying them is a separate manual step. </objective> <execution_context> Arguments parsed from `$ARGUMENTS`: - First positional: `<run-id>` or path to a `calls.ndjson` file or its …
Read the whole file at its exact version.
How to install
mdr add enricai/leerie/llm-self-heal@sha256:ef5be4a3298ff703The line at the top of this page pins the label git:20260702.2f2185f, which is what most people want. This one pins the sha256 fingerprint of the contents instead, so it survives a label being moved. Either way the exact version is written to mdr.lock, and mdr install downloads and checks those contents again, so it installs them exactly or stops with an error.
[](https://markdownregistry.com/a/art_vmgdtxtpuytqzhq4)
2 badge image fetches in 30 days. What the badge shows.
Versions
The registry has recorded 5 versions of this file, between May 28, 2026 and Jul 2, 2026, about one every 9 days.
| version | committed | commit | size | audit | |
|---|---|---|---|---|---|
| git:20260702.2f2185f latest | 2026-07-02 | 2f2185f | 12,191 B | A | view · diff |
| git:20260702.5a086eb | 2026-07-02 | 5a086eb | 11,490 B | A | view · diff |
| git:20260602.60ee283 | 2026-06-02 | 60ee283 | 11,214 B | A | view · diff |
| git:20260602.08cf96f | 2026-06-02 | 08cf96f | 11,230 B | A | view · diff |
| git:20260528.75194ec | 2026-05-28 | 75194ec | 11,192 B | A | view |
Audit of the latest version
- pass: Frontmatter block present
- pass: Frontmatter declares a name
- pass: Frontmatter declares a description
- pass: Size between 200 bytes and 200 KB (12191 bytes)
- pass: No zero-width or bidi control characters
- pass: No instruction hidden inside an HTML comment
- pass: No link to an exfiltration or paste host
- pass: No credential-shaped string
- pass: No instruction to send local credentials anywhere
- pass: No text hidden with inline styles
- pass: No prompt-injection phrasing
- pass: No curl or wget piped into a shell
- pass: No recursive delete of root, home or parent
- pass: No instruction to read or print local credentials
- pass: No base64 blob over 200 characters
- pass: No link to a raw IP address
- pass: No script tag
Source
enricai/leerie · 10 stars · license MIT · pushed 2026-10-15 · branch main
API
GET https://markdownregistry.com/api/v1/artifacts/art_vmgdtxtpuytqzhq4 GET https://markdownregistry.com/api/v1/resolve?ref=enricai/leerie/llm-self-heal GET https://markdownregistry.com/api/v1/blob/ef5be4a3298ff703e4d45a3de4eb3c57a3d2626f02447fb1aeb22b99edee1bf1
Your AI assistant does the legwork. You hear about the deals worth your word. Hand yours the standing instructions at modelranch.com and it joins the network that reads files like this one.
Guides: the SKILL.md format and its required fields · what to check before you install a skill · how to pin a skill to an exact version · where to find agent skills · keeping a team on the same skills · which agent file does what.