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 skills add StepzeroLab/research-git --skill rgit-digestgit clone --depth 1 https://github.com/StepzeroLab/research-gitWrote 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/skills/stepzerolab/research-git/rgit-digest)<a href="https://agentmods.dev/skills/stepzerolab/research-git/rgit-digest"><img src="https://agentmods.dev/badge/skills/stepzerolab/research-git/rgit-digest/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/stepzerolab/research-git/rgit-digest"><img src="https://agentmods.dev/badge/skills/stepzerolab/research-git/rgit-digest.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector warn
SkillSpector: 2 findings, up to medium
These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →
- medium Excessive Agency · line 37 Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.Fix: Add human-in-the-loop confirmation for destructive, irreversible, or high-impact operations. Never auto-execute commands that modify files, send data, or alter system state.
- medium Excessive Agency · line 44 Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.Fix: Add human-in-the-loop confirmation for destructive, irreversible, or high-impact operations. Never auto-execute commands that modify files, send data, or alter system state.
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 | $0.00076 | $0.00969 |
| Opus 5.5 | $0.00030 | $0.00388 |
| Sonnet 5.5 | $0.00015 | $0.00194 |
| Haiku 4.5 | $0.00008 | $0.00097 |
Grade A, and why
rgit-digest scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2mo ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 64 lines — stays where its author put it; the contents beside it link to each section on GitHub.
rgit-digest
Drains the history-digestion queue: deterministic staging through the rgit CLI, semantic segmentation via the capsule-segmenter subagent, non-interactive ingestion as origin=backfill capsules. Unlike live capture there is NO human approval gate — backfilled capsules are marked, filterable in recall, and bulk-removable (rgit digest clear).
Prerequisites: the repo is rgit init-ed and rgit digest scan has run (init usually does this; run it yourself if rgit digest status shows no units).
Locating the agent definitions. Same as rgit-capture: on Claude Code the plugin runtime resolves agent paths; on other CLIs resolve the plugin root from the skill symlink:
SKILL_REAL=$(realpath ~/.agents/skills/rgit-digest 2>/dev/null || readlink -f ~/.agents/skills/rgit-digest)
PLUGIN_ROOT=$(dirname "$(dirname "$SKILL_REAL")")
Process
1. Report the plan and agree on scope
Run rgit digest status --json. Tell the user: pending units, how many are dead experiments, estimated batches. Agree how many batches to run this session (default: keep going until the queue is empty or the user stops you).
2. Stage a batch
rgit digest next --batch 10 --json
Each item is {unit_id, kind, score, proposal_id, meta, diff, candidates, oversized}. meta carries commit subjects, dates, author — and for dead units the revert info (reverted_by / revert_subject / revert_date) when the death was a revert. Empty-diff units never appear (the engine skips them itself).
3. Dispatch capsule-segmenter per item (concurrently)
For each item, dispatch a subagent from agents/capsule-segmenter.md, passing proposal_id, repo_root, diff, and a history_context block built from meta: the commit subjects + dates + author, the revert info when present, oversized when true, and the note "historical diff — today's code may have refactored past it; write the resurrection guide against intent and structure". The subagent returns {"capsules": [...], "dropped": [...]}.
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.
- 2mo ago First seen · 64 lines · 76 tokens per session scan A 0c61223b66e6
rgit-digest is a skill published in the GitHub repository StepzeroLab/research-git (44 stars, last pushed 3mo ago), licensed MIT. It adds 76 tokens to every session and 969 once invoked, about $0.0003 per session on Opus 5.5. 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 2026-08-30.
rgit-digest compared with skills from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| rgit-digestStepzeroLab/research-git | Skill | 44 | 3mo ago | MIT | A | 76 | Claude Code |
| pre-compactchrisallenlane/claude-swe-workflows | Skill | 18 | 5mo ago | MIT | A | 69 | Claude Code |
| project-memorymaystudios/maxsimcli | Skill | 11 | 7mo ago | none | A | 41 | Claude Code, Codex |
| dataladK-Dense-AI/scientific-agent-skills | Skill | 56k | 4d ago | MIT | A | 147 | Claude Code |
| prYeachan-Heo/oh-my-claudecode | Skill | 41k | yesterday | MIT | A | 47 | Claude Code |
| agentmemory-agentsrohitg00/agentmemory | Skill | 30k | yesterday | Apache-2.0 | A | 46 | Claude Code |
| baby-sitlangchain-ai/open-swe | Skill | 11k | yesterday | MIT | A | 30 | Claude Code, Codex |