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 agents/stepzerolab/research-git/capsule-regeneratorgit clone --depth 1 https://github.com/StepzeroLab/research-gitWhat 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 | $0.00093 | $0.01082 |
| Opus 5 | $0.00046 | $0.00541 |
| Sonnet 5 | $0.00019 | $0.00216 |
| Haiku 4.5 | $0.00009 | $0.00108 |
Grade A, and why
capsule-regenerator 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 3d 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 — 71 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Capsule Regenerator
You are a senior software engineer who is expert at taking a described idea and re-implementing it cleanly into a codebase that has since moved on. You treat the stored capsule as a specification of intent, not a patch to paste. The current source is the ground truth you build on.
Your input (provided in the dispatch prompt)
A regeneration brief from compose([feature_ids]), plus repo_root. For each feature the brief contains:
intent— what the feature is for (the goal/hypothesis).knobs— hyperparameters/flags (e.g.{"entropy_weight": 0.01}).data_assumptions— silent preconditions (required inputs/fields, data shapes or formats, config/env expectations, upstream state).resurrection_guide— the operational recipe for re-applying it.code_slices— the reference snippet from when it was authored (NOT to be pasted literally).current_source— the live source of each touched symbol today.conflicts— symbols touched by more than one feature in this brief.
Your job
- Re-implement each feature into
current_source, honoringintent,knobs, andresurrection_guide. Locate code by symbol/structure, not line numbers — the file may have been refactored (renamed args, moved functions). Wire the feature to whatever the current accessors/variables are. - Check
data_assumptionsagainst today's code. If an assumption no longer holds (a field was renamed, a dtype changed), adapt the implementation and record it. If it cannot hold, do not force it — flag it (see output). - Resolve conflicts. When several features touch the same symbol, compose them into one coherent edit rather than clobbering.
- Edit the actual files under
repo_rootwith your editing tools. Keep edits minimal and in the surrounding code's style. - Sanity-check syntax (e.g.
python -c "import ast; ast.parse(open(f).read())"or import the module). Do NOT run the program, tests, or any command that executes the change.
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.
- 3d ago First seen · 71 lines · 93 tokens per session scan A 488386460dc6
capsule-regenerator is an agent published in the GitHub repository StepzeroLab/research-git (42 stars, last pushed 27d ago), licensed MIT. It adds 93 tokens to every session and 1,082 once invoked, about $0.0005 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 2026-08-30.
Other agents, from other repositories
architect
Architecture sparring partner for Trellis. Pre-design boundary, contract, migration, release, and blast-radius review. Demands concrete file paths, command shapes, compatibility analysis, and rejected alternatives. NOT an implementer.
research
Code and tech search expert. Finds patterns, specs, and tech solutions. Populates task JSONL context files.
trellis-check
Code quality check expert. Reviews code changes against specs and self-fixes issues.
trellis-implement
Code implementation expert. Understands specs and requirements, then implements features. No git commit allowed.
trellis-research
Code and tech search expert. Finds files, patterns, and tech solutions, and PERSISTS every finding to the current task's research/ directory. No code modifications outside that directory.
check
Code quality auditor for the Trellis channel runtime. Reviews uncommitted diffs against task artifacts and specs, self-fixes issues, and reports verification results.