bostonaholic/skills

learning-from-mistakes

Codifies the latest correction or mistake in the conversation as a short imperative rule in the project or global CLAUDE.md or AGENTS.md, after the user approves it.

View source

How to use

  1. Copy the installation command and review the source package before installing it in your own agent environment.
  2. Check the source document for required tools, credentials and supported hosts.
  3. Describe your input and desired deliverable; review the output against the source material.
npx skills add https://github.com/bostonaholic/skills --skill learning-from-mistakes

Before you start

  • Installation does not include a model subscription, API allowance or permission to access your files.
  • Repository stars and installation counts describe activity, not an independent quality or security test.
Original skill document

Rendered from the source repository. Headings, examples, code, tables, links, and referenced images are preserved.

Learn From Mistake

Turn the most recent mistake or correction in the conversation into a rule in the instructions file the agent reads, so the lesson persists across sessions. If no correction is clear, ask which one to codify.

  • The rule: one or two imperative sentences ("Do X", "Never Y"), specific

enough to change behavior. A platitude ("be careful with tests") fails.

  • Target file:
  • Project lessons (this codebase, its conventions or tooling) go in the

instructions file at the root from git rev-parse --show-toplevel: CLAUDE.md in Claude Code, AGENTS.md in Codex. Outside a repository, ask where to write.

  • Lessons that apply across projects go in ~/.claude/CLAUDE.md in Claude

Code or ~/.codex/AGENTS.md in Codex. In another host, ask which file it reads.

  • If the file is a symlink, edit the file it resolves to. If it only imports

another file (a line such as @AGENTS.md), edit the imported file.

  • Overlap: if the rule already exists, stop and say where. If a related

rule exists, refine it instead of adding a duplicate.

  • Approval: show the rule (or the old and new text of a refined rule), the

file path, and the section: an existing section that fits, or else a new ## Learned Rules section. Never propose any other new heading. Write nothing until the user approves.

Related by task

from this repository

More skills

All skills
bostonaholic
Community

auditing-complexity

Ranks where code complexity concentrates in a git repository (cyclomatic complexity, nesting, fan-out, mutable state) and scores CRAP change risk from a coverage file. Read-only. Use when asked for complexity hotspots or change risk. Not for test value; use auditing-tests.

installs
1
GitHub stars
2
Updated
Oct 20
bostonaholic
Community

composing-agent-prompts

Composes a self-contained, source-cited task prompt for another coding agent to execute in a target repository, without running the task. Use when the user asks to write, draft, or hand off a prompt, brief, or task spec for another agent. Not for system prompts; use writing-system-prompts.

installs
1
GitHub stars
2
Updated
Oct 20
bostonaholic
Community

conducting-deep-research

Conducts source-backed, goal-directed research that combines a scholarly evidence review with an institution, company, or funding map. Use when a request needs both an evidence review and an organization or funding map. Not for quick lookups or general research write-ups.

installs
1
GitHub stars
2
Updated
Oct 20
bostonaholic
Community

cutting-skills

Audits or trims an existing agent skill to its behavioral core and narrows over-broad trigger metadata. Use when the user asks to cut, simplify, shorten, de-slop, or reduce the context cost or trigger aggressiveness of a SKILL.md or skill folder. Never infer from a skill being long or edited for another purpose.

installs
1
GitHub stars
2
Updated
Oct 20