compact-task-memory
SolidProduce a lossy compaction summary of TASK_STATE.md when task memory has grown beyond a useful working size, preserving canonical decisions and recommended next step while dropping stale facts. Distinct from sync-task-state (incremental, append-only, never lossy) and state-reconcile (drift repair, no shrinking). Use when task memory has accumulated across multiple slices (5+ completed) and feels heavy, the resume cost is growing as the file scales, the current known facts list is full of resolved or routine entries, or before a session pause where a slim TASK_STATE will speed restart. Do not use when the task is still in early discovery (memory is small), the artifacts disagree across files (use state-reconcile first), an incremental sync would be sufficient (use sync-task-state), or no active task folder exists yet (run task-init first).
Install
Quality Score: 81/100
Skill Content
Details
- Author
- Mozurok
- Repository
- Mozurok/fhorja.dev
- Created
- 1 months ago
- Last Updated
- 5 days ago
- Language
- Python
- License
- MIT
Similar Skills
Semantically similar based on skill content — not just same category
capture-observation
Append a single observation, question, hypothesis, or concern to TASK_STATE.md as task memory without disrupting in-progress work or requiring a full state sync. Lean append-only; never restructures other artifacts. Use when something surfaces mid-work that should be remembered for later (during planning, implementation, or review), when you noticed something not actionable now but should not be lost, when you want to log a hypothesis to validate before closure, or when a small concern does not justify decision-interview or targeted-questions but deserves to be on record. Do not use when the observation is a canonical decision (use decision-interview or sync-task-state), invalidates the current direction (use direction-adjust), requires several artifact changes (use sync-task-state or state-reconcile), or when there is no active task folder yet (run task-init first).
im-stuck
Break the task out of a loop, confusion state, or false-progress state, and determine the fastest safe path forward. Diagnoses whether the issue is technical, workflow-related, or scope-related, then routes to the right recovery command and editor mode (Ask by default; Debug if the stuckness is a concrete technical failure; Plan if it is a phase or contract or sequence issue). Use when progress is looping or stalling, the same questions or reviews are being repeated, the wrong command or wrong editor mode may be in use, or when the user is unsure whether the problem is technical, workflow-related, or scope-related. Do not use when the next step is already clear, the task only needs normal routing via what-next, or when the task is brand-new and should start with task-init. When the real blocker is a vague prompt, use prompt-shape.
memory-compact
Compact and structure a Claude Code auto-memory store so MEMORY.md stays under the harness load cap (first 200 lines / 25KB load each session — content past that SILENTLY never loads) while every memory stays reachable. Splits the index into two tiers — hot MEMORY.md (laws/preferences/live-keys/in-flight/open-research) + cold MEMORY_archive.md (shipped/fixed/forensic/dated, recalled on demand) — and proves "done" with an integrity witness (check_memory.py) that re-derives both caps + the both-files bijection from disk, never from narration. Use when the operator says "compact memory", "MEMORY.md is too big / keeps truncating / shows as large", "tier the memory", "fix what loads for this project", or after any heavy pass that grew the index. Read-only except the memory files it edits; deletions need explicit sign-off.