Getting it into your agent
It runs from inside its repository, so the clone comes first — what it calls does not travel with the file alone.
git clone --depth 1 https://github.com/dynos-fit/dynos-worknpx agentmods add skills/dynos-fit/dynos-work/calibrationWrote 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/dynos-fit/dynos-work/calibration)<a href="https://agentmods.dev/skills/dynos-fit/dynos-work/calibration"><img src="https://agentmods.dev/badge/skills/dynos-fit/dynos-work/calibration.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 | $0.00056 | $0.01884 |
| Opus 5 | $0.00028 | $0.00942 |
| Sonnet 5 | $0.00011 | $0.00377 |
| Haiku 4.5 | $0.00006 | $0.00188 |
Grade A, and why
calibration 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 8d 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 — 151 lines — stays where its author put it; the contents beside it link to each section on GitHub.
dynos-work: Calibration
Calibrates the system's agents to your project. Generates project-specific specialists from task retrospectives, benchmarks them against generics, promotes when they outperform, archives when they regress.
If available in this repo, the deterministic runtime for registry, routing, promotion, and automatic challenger execution is:
python3 memory/agent_generator.py init-registry --root .
python3 memory/agent_generator.py register-agent <agent_name> <role> <task_type> <path> <generated_from> --root .
python3 memory/agent_generator.py auto --root .
python3 hooks/eval.py evaluate candidate.json baseline.json
python3 hooks/eval.py promote <agent_name> <role> <task_type> candidate.json baseline.json --root .
python3 hooks/bench.py run benchmarks/fixtures/<fixture>.json --root . --update-registry
python3 hooks/rollout.py benchmarks/fixtures/<rollout-fixture>.json --root . --update-registry
python3 hooks/router.py resolve <role> <task_type> --root .
python3 hooks/fixture.py sync --root .
python3 hooks/report.py --root .
python3 hooks/auto.py sync --root .
python3 hooks/auto.py run --root .
Note: the older hooks/calibrate.py and hooks/generate.py wrappers were removed in commit ae237ec; their functionality lives in memory/agent_generator.py (which exposes auto, init-registry, and register-agent subcommands directly). The dynos calibration and dynos evolve shell aliases route to it via bin/dynos.
What you do
Step 1 -- Agent Generation
Generate learned agent or skill .md files when specialization opportunities are detected. This step runs inline (no subagent spawns). Every generated runtime component must also be registered in .dynos/learned-agents/registry.json.
Prefer hooks/generate.py when you want a deterministic learned component file instead of an ad hoc markdown draft.
1a -- Generation gate
All three conditions must be true to proceed. If any is false, skip Step 1 silently.
- Sufficient data: At least 5 retrospectives with reward data (
quality_scorepresent). - Rate limit: No generation occurred in the last 3 tasks. The last generation task ID is persisted in
project_rules.mdunder the## Agent Routingsection asLast generation: {task-ID}. Compare the current task ID against the stored value; if fewer than 3 task IDs have elapsed, skip. If no stored value exists, the condition is satisfied. - Triggered Execution: This step runs when the evolve skill is invoked (typically after learn).
What ships with it
1 file beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- 8d ago First seen · 151 lines · 56 tokens per session scan A 0428aad2a693
calibration is a skill published in the GitHub repository dynos-fit/dynos-work (2 stars, last pushed yesterday), licensed MIT. It adds 56 tokens to every session and 1,884 once invoked, about $0.0003 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.
Other skills, from other repositories
ring:migrating-to-lib-observability
Migrating a Lerian Go app off lib-commons observability imports (deprecated shims or removed APIs) to lib-observability via a fixed mapping table, then bumps go.mod and validates the build; ring:backend-go applies the edits. Covers log/zap/runtime/assert, opentelemetry/tracing, HTTP middleware, context helpers, and…
ring:using-lib-streaming
Using lib-streaming, Lerian's producer-only event publication library (Kafka/SQS/RabbitMQ/EventBridge), in two modes. Sweep Mode detects DIY publishers (franz-go, sarama, amqp091, watermill, raw AWS SDK) and re-rolled manifests/breakers. Reference Mode catalogs the Builder/Emitter/Catalog facade. Companion to…
ring:using-lib-systemplane
Using lib-systemplane, the hot-reload runtime-config plane (Postgres LISTEN/NOTIFY or MongoDB change streams), in two modes. Sweep Mode detects DIY config reload (SIGHUP, fsnotify, viper, pgx LISTEN), manual tenant-scoping, hand-built admin CRUD, and v4 residue. Reference Mode catalogs client lifecycle and…
ring:applying-composition-patterns
React composition patterns that scale. Avoid boolean prop proliferation by using compound components, lifting state, and composing internals. Use when refactoring components with boolean prop proliferation, building flexible component libraries, or during architecture review. Skip for simple components with 1-2 props…
ring:using-lib-observability
Using lib-observability v1.1.0, Lerian's OpenTelemetry foundation (lib-commons, lib-systemplane, lib-streaming depend on it), in two modes. Sweep Mode detects DIY zap/slog logging, raw OTel metrics, hand-rolled redaction, and hard-coded attribute strings. Reference Mode catalogs the log, metrics, zap, redaction, and…
ring:using-outbox
Using the transactional-outbox pattern across lib-streaming (writer) and lib-commons/v5/commons/outbox (repository + relay), in two modes. Sweep Mode detects DIY outbox tables, hand-rolled relay loops, send-and-pray emits, missing WithOutboxTx wrapping, and broker calls inside DB transactions. Reference Mode catalogs…