Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add commands/omrikais/cctally/cache-syncgit clone --depth 1 https://github.com/omrikais/cctallyWhat it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00000 | $0.02369 |
| Opus 5 | $0.00000 | $0.01184 |
| Sonnet 5 | $0.00000 | $0.00474 |
| Haiku 4.5 | $0.00000 | $0.00237 |
Grade A, and why
cache-sync scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured yesterday.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 149 lines — stays where its author put it; the contents beside it link to each section on GitHub.
cache-sync
Prime or rebuild the compact accounting cache (cache.db) and independent transcript/search store (conversations.db).
Synopsis
cctally cache-sync
[--rebuild]
[--prune-orphans]
[--prune-conversations]
[--source {claude,codex,all}]
Purpose
Most commands trigger an incremental delta-ingest implicitly when they
run. Use cache-sync when you want to:
- Force the cache up-to-date now (so the next interactive command is fast).
- Rebuild from scratch after deleting
cache.db, schema changes, or pricing-dict edits where you want a clean re-derivation (note: pricing edits don't actually require a rebuild — cost is computed at query time, not stored). - Recover a corrupt
cache.dborconversations.dbwithout moving or unlinking SQLite files by hand. Positively confirmed corruption is preserved for forensics, the complete main/WAL/SHM family is quarantined after live readers drain, and the requested source scope is re-ingested. The default--source allrecovery rebuilds both transcript providers. - Prune cache rows left behind by session directories that were removed from disk (for example a deleted git worktree), without paying for a full rebuild — see
--prune-orphansbelow. - Limit work to one source (Claude or Codex) when the other half is large.
Options
| Flag | Description |
|---|---|
--rebuild |
Drop all cached entries and re-ingest from scratch. It first checks an existing transcript store and safely recovers confirmed corruption. Waits up to 30s for each provider lock; each transcript-provider phase has a 30-minute no-progress ceiling and exits non-zero if incomplete (see Notes). |
--prune-orphans |
Remove cache rows for source files no longer on disk, without a full rebuild (Claude cache only). |
--prune-conversations |
Prune conversation transcripts older than conversation.retention_days (default 90) right now, without a full rebuild. Reports the rows removed per provider. See --prune-conversations below. |
--source {claude,codex,all} |
Which ingest half to sync/rebuild. Default all. |
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- yesterday First seen · 149 lines · 0 tokens per session scan A 2b0417a7111c
cache-sync is a command published in the GitHub repository omrikais/cctally (5 stars, last pushed 3d ago), licensed Apache-2.0. It costs nothing until one of its globs matches a file; then it loads 2,369 tokens. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other commands, from other repositories
checklist
Generate a custom checklist for the current feature based on user requirements.
clarify
Identify underspecified areas in the current feature spec by asking up to 5 highly targeted clarification questions and encoding answers back into the spec.
specify
Create or update the feature specification from a natural language feature description.
analyze
Perform a non-destructive cross-artifact consistency and quality analysis across spec.md, plan.md, and tasks.md after task generation.
converge
Assess the current codebase against the feature's spec, plan, and tasks, then append any remaining unbuilt work as new tasks to tasks.md so implement can complete it.
implement
Execute the implementation plan by processing and executing all tasks defined in tasks.md.