← Back to Skills Marketplace
Claw History
by
BlueBirdBack ✨
· GitHub ↗
· v1.0.0
599
Downloads
0
Stars
1
Active Installs
1
Versions
Install in OpenClaw
/install claw-history
Description
Provide a chronological history of all actions the agent has taken from the beginning (birth) until now. Use when the user asks for full lifetime timeline/ac...
README (SKILL.md)
Claw History
Return the most complete, auditable lifetime timeline possible.
Core Rule
Do not default to current-session history. Reconstruct from earliest available evidence to now.
Required Data Sources (in this order)
- Earliest and recent
memory/YYYY-MM-DD*.mdfiles (find first/last dates). MEMORY.mdfor long-term milestones.- Available session inventory/history (main + sub-agent sessions when accessible).
- Current conversation/tool logs.
If any source is unavailable, explicitly state that gap.
Workflow
- Determine earliest known timestamp (“birth marker”) from records.
- Collect confirmed actions across all available periods.
- De-duplicate overlapping entries from memory/session logs.
- Sort strictly oldest → newest.
- Mark each entry as:
- Confirmed (direct evidence), or
- Inferred (indirect/partial evidence).
- Add a final coverage summary with what was and was not visible.
Output Format
- Birth marker: first known timestamp + source
- Scope covered: date range + sources checked
- Chronological history (oldest → newest):
YYYY-MM-DD HH:MM (TZ) — [Confirmed|Inferred] Action — Result
- Gaps / limits: missing windows, inaccessible logs, uncertainty
- Confidence: high / medium / low
Quality Rules
- Never invent timestamps or actions.
- Prefer evidence-backed entries over narrative summaries.
- If exact time is unknown, use best-known granularity (date/session order) and label it.
- Include source pointers when possible (file + line or log reference).
- If user requested “all actions,” include both major milestones and notable operational actions.
Usage Guidance
This skill will try to read your agent's local memory files and logs to reconstruct a complete timeline. That behavior is expected for a history tool, but you should: (1) confirm you are comfortable with the agent reading workspace/memory/*.md, MEMORY.md, session inventories, and any command logs; (2) only enable the recommended hooks (session-memory, command-logger) if you want persistent, auditable logging — enabling them increases retained telemetry; (3) consider running in an isolated workspace if logs contain sensitive data; and (4) verify there are no unexpected network endpoints or credential requests (none are declared). If you want tighter limits, ask the agent to restrict sources (e.g., only current session) before running the skill.
Capability Analysis
Type: OpenClaw Skill
Name: claw-history
Version: 1.0.0
The `claw-history` skill is designed to provide a chronological history of the agent's actions, which is a legitimate auditing and debugging function. The `SKILL.md` instructs the agent to access its own internal memory files (`memory/YYYY-MM-DD*.md`, `MEMORY.md`) and logs, which is necessary for its stated purpose. The `README.md` provides instructions for the user to enable OpenClaw hooks (`session-memory`, `command-logger`) to improve history completeness, which are standard OpenClaw CLI commands and not malicious code. There is no evidence of data exfiltration, malicious execution, persistence, obfuscation, or prompt injection attempts to manipulate the agent into harmful or unauthorized actions. All instructions and data access are consistent with the skill's transparent auditing goal.
Capability Assessment
Purpose & Capability
The name/description (produce a lifetime action timeline) aligns with the resources the skill asks to read: memory/YYYY-MM-DD*.md, MEMORY.md, session inventories, and conversation/tool logs. These data sources are reasonable for producing a full-history report.
Instruction Scope
SKILL.md explicitly directs the agent to scan multiple local data sources (workspace memory files, session inventories, sub-agent sessions when accessible, and tool logs). That is coherent with the stated goal but means the skill will attempt to read potentially sensitive local files and cross-session data. It correctly requires the skill to report gaps when sources are unavailable.
Install Mechanism
Instruction-only skill with no install spec or external downloads; lowest install risk. No code files are included, so nothing is written to disk by an installer.
Credentials
No environment variables, credentials, or external endpoints are requested. The only access needed is read access to local memory/log files, which is proportional to building a lifetime timeline.
Persistence & Privilege
always:false and no indications the skill modifies other skills or system-wide settings. The README recommends enabling optional hooks (session-memory, command-logger) for better coverage — enabling those is an explicit user action and not required by the skill itself.
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install claw-history - After installation, invoke the skill by name or use
/claw-history - Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.0
Initial public release
Metadata
Frequently Asked Questions
What is Claw History?
Provide a chronological history of all actions the agent has taken from the beginning (birth) until now. Use when the user asks for full lifetime timeline/ac... It is an AI Agent Skill for Claude Code / OpenClaw, with 599 downloads so far.
How do I install Claw History?
Run "/install claw-history" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is Claw History free?
Yes, Claw History is completely free (open-source). You can download, install and use it at no cost.
Which platforms does Claw History support?
Claw History is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created Claw History?
It is built and maintained by BlueBirdBack ✨ (@bluebirdback); the current version is v1.0.0.
More Skills