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/record-creditgit clone --depth 1 https://github.com/omrikais/cctallyWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/commands/omrikais/cctally/record-credit)<a href="https://agentmods.dev/commands/omrikais/cctally/record-credit"><img src="https://agentmods.dev/badge/commands/omrikais/cctally/record-credit.svg" alt="Measured on agentmods" height="20"></a>What 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 5.1 | $0.00000 | $0.02824 |
| Opus 5 | $0.00000 | $0.01412 |
| Sonnet 5 | $0.00000 | $0.00565 |
| Haiku 4.5 | $0.00000 | $0.00282 |
Grade A, and why
record-credit 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 5d ago.
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 — 136 lines — stays where its author put it; the contents beside it link to each section on GitHub.
record-credit
Record an in-place weekly (7d) credit — a case where Anthropic lowered your 7-day usage counter mid-window without a clean reset to zero — and repair the local data so reports and the statusline read the credited value instead of the stale pre-credit high-water mark.
Synopsis
cctally record-credit
--to TO
[--from FROM]
[--at AT]
[--week WEEK]
[--dry-run] [--yes] [--force] [--json]
Purpose
cctally already auto-detects two shapes of an in-place weekly credit during record-usage: a big drop (≥25 percentage points, fired immediately) and a reset-to-zero (post-value ≤1%, debounced). A real incident on 2026-06-19 fell between the cracks: the weekly counter changed 46% → 31%, a 15pp drop whose post-value is nowhere near zero. That is below the 25pp threshold and is not a zero-collapse, so neither auto-detector fires — and the write-site monotonicity clamp then treats the lower 31% reading as a lagging-replica regression and refuses to store it, leaving every display pinned at the stale 46%.
The 25pp threshold is deliberate: a slightly-behind API replica reports a marginally-lower number, and a 15pp drop is fundamentally ambiguous from percent-shape alone, so lowering the threshold would reintroduce exactly the false positives it guards against. record-credit is the deterministic, user-asserted escape hatch: you state the credit explicitly (zero false-positive risk), and the command repairs the local data so every current-week surface reads the credited value.
Same week — no re-anchor. A partial credit is not a reset: the week keeps its original boundaries (e.g. 2026-06-13 → 2026-06-20); only the running 7d percent steps down (46 → 31, then climbs again as usage resumes). record-credit records this by writing a row to weekly_credit_floors — a clamp floor for the current week — and does not write a week_reset_events row, so the window-resolution code never re-anchors the week to the credit moment (which would otherwise show a spurious "new week" and corrupt the forecast rate). The four MAX-clamp sites that derive the current 7d percent (the statusline 7d high-water mark, the record-usage write-site monotonic clamp, the --from default helper, and project's per-week usage) consult the union of week_reset_events + weekly_credit_floors and floor the displayed value to the post-credit reading while the window stays put.
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.
- 5d ago First seen · 136 lines · 0 tokens per session scan A 1fdde85664e0
record-credit is a command published in the GitHub repository omrikais/cctally (5 stars, last pushed yesterday), licensed Apache-2.0. It costs nothing until one of its globs matches a file; then it loads 2,824 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.