codeops_stats

A command for viewing CodeOps quality measurements, such as event counts, agent runs, acceptance rates, and durations.

In plain words
What is it for?
It helps review quality by agent, project, event, delivery stage, or investigation period, and identify missing event records.
Why use it?
It turns recorded tool-use events into compact tables, so you do not need to inspect raw telemetry files.

Command

Install

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.

agentmods
npx agentmods add commands/blendsdk/claude-codeops/codeops_stats
Clone the repo
git clone --depth 1 https://github.com/blendsdk/claude-codeops
Per session 83 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 603 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5 $0.00083 $0.00603
Opus 5 $0.00042 $0.00302
Sonnet 5 $0.00017 $0.00121
Haiku 4.5 $0.00008 $0.00060

Measured yesterday against content hash ecf4d1a9b901, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

codeops_stats 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.

commands/codeops_stats.md · 43 lines

What it actually says

codeops_stats — relay the telemetry tables

The utility at "${CLAUDE_PLUGIN_ROOT}/scripts/codeops-events.sh" is the ONLY reader of the telemetry file (~/.claude/codeops-telemetry/events.jsonl). This command runs it and relays its output verbatim — never open, grep, or summarize events.jsonl yourself; the utility pre-aggregates to a table of at most ~40 lines precisely so raw events stay out of context.

Mapping the ask to an invocation

User asks for Run
overall picture (default) codeops-events.sh stats
per-agent rates codeops-events.sh stats --by agent
per-lens rates codeops-events.sh stats --by lens
per-project / per-event counts codeops-events.sh stats --by project / --by event
planned vs. verified, first-pass rate, rework codeops-events.sh stats --by delivery
runtime ambiguity by owning stage, resume accuracy codeops-events.sh stats --by drift
whether delegated design decisions resolved or escalated codeops-events.sh stats --by design
a time window ("last 2 weeks") add --since 14d
one project only add --project <name>
emission gaps ("reviews without rulings") codeops-events.sh gaps [--since <Nd>]

Pass $ARGUMENTS through when the user already typed flags.

Relay rules

  • Print the utility's table as-is (a fenced code block keeps the alignment). Add at most a one-line reading of what stands out — no re-derivation, no editorializing beyond the numbers.
  • no events recorded → explain the likely causes: telemetry off (CODEOPS_TELEMETRY=0, or telemetry: off in the repo's quality block), jq missing, or simply nothing has run yet.
  • For threshold judgments ("is this acceptance rate bad?") point to /codeops_retro — that command owns the thresholds; this one only reports.
Changes

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.

  1. yesterday First seen · 43 lines · 83 tokens per session scan A ecf4d1a9b901

Subscribe to this mod's changes

codeops_stats is a command published in the GitHub repository blendsdk/claude-codeops (4 stars, last pushed 1mo ago), licensed MIT. It adds 83 tokens to every session and 603 once invoked, about $0.0004 per session on Opus 5. 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.