Borrowing it
Nothing to install: this file belongs to jongwony/epistemic-protocols. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/jongwony/epistemic-protocols/main/.claude/skills/realize/SKILL.mdgit clone --depth 1 https://github.com/jongwony/epistemic-protocolsWrote 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/jongwony/epistemic-protocols/realize)<a href="https://agentmods.dev/skills/jongwony/epistemic-protocols/realize"><img src="https://agentmods.dev/badge/skills/jongwony/epistemic-protocols/realize/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/jongwony/epistemic-protocols/realize"><img src="https://agentmods.dev/badge/skills/jongwony/epistemic-protocols/realize.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
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 | $0.00086 | $0.02900 |
| Opus 5.5 | $0.00034 | $0.01160 |
| Sonnet 5.5 | $0.00017 | $0.00580 |
| Haiku 4.5 | $0.00009 | $0.00290 |
Grade A, and why
realize 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 — 226 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Type Realization
Measure whether one named protocol skill's declared contract is realized during an
actual run. The first argument is required; /realize inquire selects only the
registered inquire cases and graders. An unknown or omitted target fails before setup.
Purpose
A protocol's formal blocks are runtime-normative: TYPES, PHASE TRANSITIONS, TOOL
GROUNDING and the Rules type the prose and carry the operational contract. Static
checks establish that a SKILL.md contains those blocks and enforce a bounded set of
structural invariants. Nothing establishes that a run follows their transitions.
This skill supplies evidence for that gap. It drives Claude Code or Codex across a
matrix of models and treatments, captures the JSONL trace, grades the deterministically
observable subset, and names the transcript judgments still owed. A complete automatic
cell can establish that collection occurred, the declared branch reached Stop, or
the zero-uncertainty branch reached Proceed; semantic ordering and rendered type
coverage remain manual until their specified judges are built.
Judgment boundary
The object of judgment is the formal transition trace:
X → context sufficiency → evidence collection as needed → residual deficit
→ formal branch → Stop | Proceed
Every score attaches to a node or edge in that trace. Once the selected branch is
observable, judgment ends. A plan, implementation, analysis, or other downstream
artifact is evidence that Proceed occurred; its quality, correctness, completeness,
and usefulness are outside /realize. Likewise, an unchanged substrate is evidence of
Stop only where the case granted the capability to proceed and the formal branch
required stopping. references/grader-design.md carries the witness design and the
separate oracle that artifact-quality evaluation would require.
When it applies
Reach for this after changing a protocol's phase structure, terminal conditions, gate definition, or answer types — the places where a contract can become unrealizable without any static check noticing. Reach for it also when adopting a new model, where the question is whether the contract still holds on it.
What ships with it
60 files 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.
- evals/conduct-map-gate/case.yaml 420 B
- evals/conduct-map-gate/graders/contrary-grounds-shown.md 2.4 KB
- evals/conduct-map-gate/graders/method-written-out.md 4.1 KB
- evals/conduct-map-gate/graders/turn-ends-at-gate.md 3.3 KB
- evals/conduct-map-gate/prompt.md 713 B
- evals/conduct-relay/case.yaml 427 B
- evals/conduct-relay/graders/map-relayed-before-dispatch.md 4.4 KB
- evals/conduct-relay/graders/proceed-observed.md 3.1 KB
- evals/conduct-relay/prompt.md 2.0 KB
- evals/conduct-scaffold.sh 9.2 KB runs code
- evals/conduct-taking-with-change/case.yaml 701 B
- evals/conduct-taking-with-change/graders/map-relayed-before-dispatch.md 4.5 KB
- evals/conduct-taking-with-change/graders/relayed-not-gated.md 3.4 KB
- evals/conduct-taking-with-change/graders/turn-ends-at-gate.md 741 B
- evals/conduct-taking-with-change/oracle.md 2.9 KB
- evals/conduct-taking-with-change/prompt.md 735 B
- evals/conduct-taking-with-change/reply-1.md 199 B
- evals/elicit-aporia/case.yaml 952 B
- evals/elicit-aporia/graders/activation-silent.md 1.1 KB
- evals/elicit-aporia/graders/adoption-valid.md 2.0 KB
- evals/elicit-aporia/graders/basis-faithful.md 2.9 KB
- evals/elicit-aporia/graders/continuation-authorized.md 1.2 KB
- evals/elicit-aporia/graders/gate-stopped.md 2.1 KB
- evals/elicit-aporia/graders/person-closes.md 1.5 KB
- evals/elicit-aporia/graders/record-cites-person.md 2.9 KB
- evals/elicit-aporia/graders/residual-and-dissent-kept.md 2.1 KB
- evals/elicit-aporia/graders/skill-fired.md 814 B
- evals/elicit-aporia/graders/substrate-read.md 3.0 KB
- evals/elicit-aporia/graders/turn-ends-at-gate.md 1.1 KB
- evals/elicit-aporia/graders/value-stands.md 1.6 KB
- evals/elicit-aporia/oracle.md 6.4 KB
- evals/elicit-aporia/prompt.md 411 B
- evals/elicit-axis-determined/case.yaml 753 B
- evals/elicit-axis-determined/graders/closes-then-proceeds.md 2.4 KB
- evals/elicit-axis-determined/graders/first-surface-faithful.md 2.6 KB
- evals/elicit-axis-determined/graders/record-honest.md 1.6 KB
- evals/elicit-axis-determined/graders/skill-fired.md 529 B
- evals/elicit-axis-determined/oracle.md 1.5 KB
- evals/elicit-axis-determined/prompt.md 1.5 KB
- evals/elicit-scaffold.sh 12 KB runs code
- evals/grasp-adjudicable/case.yaml 916 B
- evals/grasp-adjudicable/graders/closes-on-user-word.md 2.1 KB
- evals/grasp-adjudicable/graders/does-not-test.md 2.2 KB
- evals/grasp-adjudicable/graders/map-grounded.md 3.0 KB
- evals/grasp-adjudicable/graders/refutes-in-one-round.md 3.8 KB
- evals/grasp-adjudicable/graders/result-meets-both.md 3.2 KB
- evals/grasp-adjudicable/graders/stops-for-user.md 2.4 KB
- evals/grasp-adjudicable/graders/target-preserved.md 1.8 KB
- evals/grasp-adjudicable/graders/target-read-first.md 1.5 KB
- evals/grasp-adjudicable/oracle.md 4.6 KB
- evals/grasp-adjudicable/prompt.md 430 B
- evals/grasp-adjudicable/reply-1.md 439 B
- evals/grasp-adjudicable/reply-2.md 415 B
- evals/grasp-adjudicable/reply-3.md 169 B
- evals/grasp-adjudicable/reply-4.md 203 B
- evals/grasp-adjudicable/reply-5.md 100 B
- evals/grasp-unattachable/case.yaml 918 B
- evals/grasp-unattachable/graders/closes-on-user-word.md 2.1 KB
- evals/grasp-unattachable/graders/does-not-test.md 2.2 KB
- evals/grasp-unattachable/graders/map-grounded.md 3.0 KB
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 Changed · -4 lines 78f96a9d5fc0
- 14d ago Changed · +2 lines e9598bf4bbf6
- 17d ago Changed · +4 lines 6ca5abae65c9
- 19d ago Changed · +3 lines 192d441ef163
- 20d ago Changed · +2 lines 0513046b9e92
- 23d ago Changed 7325a88c6339
- 26d ago Changed · +7 lines 1a4873a0aa2f
- 28d ago Changed · +23 lines d500af2da9b7
- 29d ago Changed · +4 lines 29518f770aea
- 1mo ago First seen · 185 lines · 86 tokens per session scan A 96d44a14c448
realize is a skill published in the GitHub repository jongwony/epistemic-protocols (174 stars, last pushed yesterday), licensed MIT. It adds 86 tokens to every session and 2,900 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.
Other skills, from other repositories
flow-next-features
Seed or maintain the committed user-POV drive map at .flow/features/ so QA and drive reuse how a user reaches each feature.
taiyi-dev
A software-development stage that implements planned tasks using test-driven development, or TDD: write a failing test, make it pass, then improve the code. It checks the task plan, dependencies, file boundaries, and required completion evidence.
flow-next-qa
Live-app QA pass derived from the spec. Drives the running app, files P0/P1/P2 findings with evidence, emits a YES or NO qaverdict receipt.
taiyi-test
A project workflow skill for verifying an implementation and producing a TEST.md record. TDD means writing a failing test, implementing the change, and then making the test pass; this skill checks that process and other regression cases.
maestro-work
The contract every playbook shares - the brief a writer receives, the rules it writes by, the handback it returns, and the store record the holder keeps. Reuse existing checks and add tests only for concrete uncovered behavior or risk.
coding-guard
Compact coding guidelines for the explyt/spring-plugin repository - covers process rules (think first, simplicity, surgical changes), module layout, IntelliJ Platform threading and PSI rules, inspection and test conventions, message bundles, statistics tracking, and PR hygiene. You must call this skill before any code…