agentleFS
Sign inStart free
yaokai123yaokai123/AgentPaw/skills/swebench-recovery/SKILL.md

swebench-recovery

Skill1 starsChanged 3 months agoApache-2.0View on GitHub

Recover an empty SWE-bench run with a fresh, edit-first bounded pass.

The file

We split this skill into the rules it gives an agent

6 rules. About 0 are common, 5 are distinctive, and 1 is very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈5

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 1

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Start from the supplied recovery plan and inspected source excerpt.

    line 13
  2. Very specific:

    If exact editing fails, retry once with newline-tolerant edit_file using the context already gathered.

    line 16
  3. Distinctive:

    Use the final turns for a focused post-patch regression test or direct executable reproduction.

    line 17
  4. Distinctive:

    Prefer a small behaviorally complete patch over more investigation.

    line 18
  5. Distinctive:

    Before hand-writing a parser, graph algorithm, ordering primitive, cache, retry loop, or compatibility shim, search the repository for an …

    line 19
  6. Distinctive:

    Do not use hidden tests, gold patches, subagents, external agents, or network services.

    line 20

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

SWE-bench Recovery

The previous pass explored but left no patch. A tool-free planner may provide a bounded owning-function and edit plan. Work only inside the provided repository.

  1. Start from the supplied recovery plan and inspected source excerpt. Do not restart repository exploration.
  2. Your first tool action must create the candidate edit with edit_file or write_file.
  3. Repository reads and searches are unavailable for the entire recovery pass. Use only edit_file, write_file, or a focused validation command.
  4. If exact editing fails, retry once with newline-tolerant edit_file using the context already gathered.
  5. Use the final turns for a focused post-patch regression test or direct executable reproduction. A successful reproduction before editing proves only the baseline bug, not the fix.
  6. Prefer a small behaviorally complete patch over more investigation.
  7. Before hand-writing a parser, graph algorithm, ordering primitive, cache, retry loop, or compatibility shim, search the repository for an existing utility and its established exception/error contract.
  8. Do not use hidden tests, gold patches, subagents, external agents, or network services.
  9. Finish only after leaving a non-empty Git diff.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

Check your setup, free