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...

Security Scan

The skill's requirements and runtime instructions match its stated purpose — it reconstructs an agent timeline by reading local memory and log files — it asks for broad local read access but requires no installs, credentials, or external endpoints.

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.

Assessment

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.

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)

1. Earliest and recent memory/YYYY-MM-DD*.md files (find first/last dates).
2. MEMORY.md for long-term milestones.
3. Available session inventory/history (main + sub-agent sessions when accessible).
4. Current conversation/tool logs.

If any source is unavailable, explicitly state that gap.

Workflow

1. Determine earliest known timestamp (“birth marker”) from records.
2. Collect confirmed actions across all available periods.
3. De-duplicate overlapping entries from memory/session logs.
4. Sort strictly oldest → newest.
5. Mark each entry as:
   • Confirmed (direct evidence), or
   • Inferred (indirect/partial evidence).
6. 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.