Not in the top 1,000 by GitHub stars
pskoett/self-improving-agent
Self-improvement skill for OpenClaw. It captures learnings, errors, and feature requests to support continuous improvement across sessions. It is a agent knowledge skill made for the skill marketplace
History
Review weekly changes to SKILL.md before updating. Dates mark the weeks in which we observed each version; checks apply to the recorded commits.
View weekly changes
self-improving-agent/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Network or broad shell → After Network or broad shell
e958afb5bf69 → e958afb5bf69
Diff text is unavailable under the recorded licenses. Line counts and the GitHub link remain available.
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Network or broad shell
First observation → e958afb5bf69
No earlier version was observed for comparison.
790GitHub stars
7stars in 7 days
1skill in this repository
5commits in 30 days
Heat
4.7- Star growth, last 7 daysNot available
- Star growth, last 24 hoursNot available
- Rank movement, last 7 daysNot available
- Commits, last 30 daysNot available
- Accounts asking about it through MCP, last 7 daysNot available
- Hot votes this weekNot available
Heat never changes the order by stars, and sponsorship never touches it. How heat works
Stars over time
Individual skills in this repository
- self-improving-agent
self-improving-agent
How it works
self-improving-agent
When your agent uses it
Captures and maintains learnings, errors, and corrections. Use when a command fails, the user corrects an assumption, a capability is missing, knowledge is outdated, or a better approach is found. Review relevant saved guidance before major tasks, reuse, and promotion; revalidate it according to what could change and how quickly.
What it loads into context
Frontmatter
88 tokensAlways available to the agent
SKILL.md
6,058 tokensWhen the skill is activated
Reference documents
16,572 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- First-Use Initialisation
- Quick Reference
- OpenClaw Setup
- Logging Format
- Maintenance: Cause-Aware Context Decay
- ID Generation
- Resolving Entries
- Promoting to Workspace Memory
- Pattern-Key Taxonomy
- Recurring Pattern Detection
- Simplify & Harden Feed
- Periodic Review
- Detection Triggers
- Priority Guidelines
- Area Tags
- Best Practices
- Gitignore Options
- Upgrading & Uninstalling
- Skill Extraction (Approval Required)
Scripts executed separately
hooks/openclaw/handler.jshooks/openclaw/handler.test.jshooks/openclaw/handler.tsscripts/extract-skill.sh
Files
CHANGELOG.mddoc10,924 bytesSKILL.mdcore26,653 bytesSKILL.mdcore26,653 bytesSKILL.mdcore26,653 bytesSKILL.mdcore26,653 bytesSKILL.mdcore26,653 bytesSKILL.mdcore26,653 bytesSKILL.mdcore26,653 bytesSKILL.mdcore26,653 bytesSKILL.mdcore26,653 bytesSKILL.mdcore26,653 bytesSKILL.mdcore26,653 bytesSKILL.mdcore26,653 bytesSKILL.mdcore26,653 bytesSKILL.mdcore26,653 bytesSKILL.mdcore26,653 bytesassets/ERRORS.mddoc1,819 bytesassets/FEATURE_REQUESTS.mddoc1,654 bytesassets/LEARNINGS.mddoc3,367 bytesassets/SKILL-TEMPLATE.mddoc5,199 byteshooks/openclaw/HOOK.mddoc4,484 byteshooks/openclaw/handler.jsruns19,306 byteshooks/openclaw/handler.test.jsruns18,581 byteshooks/openclaw/handler.tsruns20,834 bytesreferences/examples.mddoc14,876 bytesreferences/maintenance-evals.mddoc7,812 bytesreferences/openclaw-integration.mddoc16,884 bytesreferences/uninstall.mddoc4,358 bytesscripts/extract-skill.shruns5,293 bytes
No excerpt is shown for this license. Read the source on GitHub.
View the complete source on GitHubSkills included
- self-improving-agent
Repository facts
- License
- No license detected
- Created
- Jan 31, 2026
- Last pushed
- Oct 22, 2026
- Forks
- 116
- Open issues
- 1
- Project activity
- Active
Rank over the last 12 weeks
Now #999 · best #664 · down 209 in 30 days
- #876 on Oct 20
- #886 on Oct 21
- #859 on Oct 22
- #874 on Oct 23
- #824 on Oct 24
- #999 since Oct 25
44 earlier periods
- #668 on Aug 2
- #664 on Aug 7
- #667 on Aug 8
- #672 on Aug 9
- #695 on Aug 10
- #707 on Aug 11
- #699 on Aug 12
- #694 on Aug 13
- #693 on Aug 14
- #668 on Aug 15
- #686 on Aug 16
- #712 on Aug 21
- #678 on Aug 22
- #711 on Aug 23
- #735 on Aug 24
- #743 on Aug 25
- #741 on Aug 27
- #764 on Aug 28
- #747 on Aug 31
- #761 on Sep 1
- #745 on Sep 2
- #749 on Sep 3
- #729 on Sep 4
- #759 on Sep 7
- #762 on Sep 9
- #776 on Sep 10
- #748 on Sep 11
- #778 on Sep 14
- #764 on Sep 15
- #775 on Sep 16
- #777 on Sep 17
- #759 on Sep 18
- #738 on Sep 19
- #790 on Sep 23
- #773 on Sep 27
- #824 on Sep 29
- #758 on Oct 1
- #757 on Oct 3
- #786 on Oct 4
- #824 on Oct 7
- #862 on Oct 8
- #828 on Oct 9
- #840 on Oct 15
- #828 on Oct 17
Before you install
Network or broad shellAutomated signals, not a guarantee
Checked Nov 4, 2026
Not fully analyzed: some files were too long, unreadable or missing. This never lowers the level.
Show 14 notes
self-improving-agent/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzedself-improving-agent/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzedself-improving-agent/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzedself-improving-agent/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzedself-improving-agent/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzedself-improving-agent/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzedself-improving-agent/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzedself-improving-agent/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzedself-improving-agent/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzedself-improving-agent/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzedself-improving-agent/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzedself-improving-agent/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzedself-improving-agent/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzedself-improving-agent/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzed
- No broad tool permissions
- No dangerous shell commands
- 13 network referencesin 4 files: self-improving-agent/CHANGELOG.md:3 self-improving-agent/SKILL.md:10 self-improving-agent/hooks/openclaw/HOOK.md:13 and 1 more
- No hidden or obfuscated text
- No instruction overrides
- 4 executable scriptsin 4 files: self-improving-agent/hooks/openclaw/handler.js:1 self-improving-agent/hooks/openclaw/handler.test.js:1 self-improving-agent/hooks/openclaw/handler.ts:1 and 1 more
- 1 reference to secretsin 1 file: self-improving-agent/hooks/openclaw/handler.test.js:227
- No requests to vote in Hot
Show all 18 signals by file
self-improving-agent/CHANGELOG.md
- Network accessline 3
All notable changes to this skill are documented here, including anything an agent or user must do when upgrading. Format follows [Keep a Changelog](https://keepachangelog.com/); versions follow - Network accessline 4
agent or user must do when upgrading. Format follows [Keep a Changelog](https://keepachangelog.com/); versions follow [SemVer](https://semver.org/) and match `version` in `SKILL.md` frontmatter. - Network accessline 5
[Keep a Changelog](https://keepachangelog.com/); versions follow - Network accessline 6
[SemVer](https://semver.org/) and match `version` in `SKILL.md` frontmatter. - Network accessline 136
multi-agent setup/support sections in `SKILL.md` and `references/uninstall.md`. For other agents, use the original multi-agent version: https://github.com/pskoett/pskoett-ai-skills. - Network accessline 138
version: https://github.com/pskoett/pskoett-ai-skills.
self-improving-agent/SKILL.md
- Network accessline 10
Log learnings and errors to markdown files for continuous improvement. Agents can later process these into fixes, and important learnings get promoted to workspace memory. This version of the skill is built for OpenClaw only — for other age - Network accessline 70
git clone https://github.com/pskoett/self-improving-agent.git /tmp/self-improving-agent-repo - Network accessline 74
Remade for openclaw from original repo : https://github.com/pskoett/pskoett-ai-skills - https://github.com/pskoett/pskoett-ai-skills/tree/main/skills/self-improvement
self-improving-agent/hooks/openclaw/HOOK.md
- Network accessline 13
callback. OpenClaw's typed plugin hooks expose `after_tool_call`; this hook retains session-end detection. See the separate [plugin hook catalog](https://docs.openclaw.ai/plugins/hooks/reference). - Network accessline 15
[plugin hook catalog](https://docs.openclaw.ai/plugins/hooks/reference).
self-improving-agent/hooks/openclaw/handler.js
- Executable scriptline 1
Executable file
self-improving-agent/hooks/openclaw/handler.test.js
self-improving-agent/hooks/openclaw/handler.ts
- Executable scriptline 1
Executable file
self-improving-agent/references/openclaw-integration.md
self-improving-agent/scripts/extract-skill.sh
- Executable scriptline 1
#!/bin/bash
How the signals fit together
Facts from a static read of the files at the pinned commit. They describe what the code does, not why.
Some files could not be read; the facts below still hold.
Signal confidence: 1 fact
- Network access in code of the SKILL.mdself-improving-agent/SKILL.md:70High confidence
Works with
| Agent | Status | Evidence | Install to |
|---|---|---|---|
| Claude Code | Compatible | Standard SKILL.md self-improving-agent/SKILL.md | .claude/skills/<skill-name>/ |
| Codex | Compatible | Standard SKILL.md self-improving-agent/SKILL.md | .agents/skills/<skill-name>/ |
| Cursor | Compatible | Standard SKILL.md self-improving-agent/SKILL.md | .cursor/skills/<skill-name>/ |
| Gemini CLI | Compatible | Standard SKILL.md self-improving-agent/SKILL.md | .gemini/skills/<skill-name>/ |
| GitHub Copilot | Compatible | Standard SKILL.md self-improving-agent/SKILL.md | .github/skills/<skill-name>/ |
| OpenCode | Compatible | Standard SKILL.md self-improving-agent/SKILL.md | .opencode/skills/<skill-name>/ |
- Native The repository ships a manifest, folder or frontmatter field for this agent.
- Compatible A standard SKILL.md that this agent can read.
- Unknown No evidence either way.
Derived automatically from paths, manifests and frontmatter. Not tested on each agent.
