A community-developed accountability skill becomes a widely adopted pattern for constraining agent actions in production harnesses.

state: seedheat: lowuncertainty: mediumconvergesscott: highagent-accountability agent-skills agent-harnesses accountability-skillkenm47hank2000

What is this?

The case centers on 'Accountability.md' — a community-developed skill for agent harnesses that aims to constrain agent actions before they execute, released via Show HN by contributors kenm47 and hank2000. The supplied web results do not directly confirm this specific release; they surface a different 'SkillHarness' research project from Zhejiang University (arXiv June 2026) that embeds safety guards in skills, a corporate 'Harness' LinkedIn post on pre-action accountability, and general industry discourse on agent governance. The evidence titles assert a Show HN launch of Accountability.md as a skill for agent harnesses, but the search snippets do not contain that announcement or its adoption signals.

Why it matters to Scott

A community-released accountability skill for agent harnesses directly implements Scott's load-bearing architectural position: governance must live in the execution path as deterministic, pre-action gates (Guardrail Illusion, Constraint Paradigm, Decision Authority Infrastructure, Zero Trust for Decisions, The Governance Stack), not as post-hoc audit or probability barriers. This skill embodies the 'skill-as-guardrail' pattern within the Skills and Workflows framework — a portable, reusable enforcement module that harnesses can compose. If it gains adoption, it becomes dated receipts that the decision-time authority layer Scott diagnosed as 'completely unbuilt' is being built by the community, not just theorised.

ip:concept.guardrail-illusionip:framework.the-governance-stackip:framework.decision-authority-infrastructureip:concept.runtime-governanceip:concept.constraint-paradigmip:concept.zero-trust-for-decisionsip:source.compliance-cosplayip:concept.skills-and-workflowsip:concept.model-plus-harness-benchmark-unitip:framework.micro-agents-architecturedev:concept.padded-cell-agent-architecturedev:concept.cheap-model-front-doordev:concept.deterministic-agent-control-planedev:technology.typesafe-jevdev:concept.validation-gated-llm-extractiondev:concept.propose-finalise-gatedev:concept.review-until-clear-loopradar:acs-local-skill-risk-catalogradar:addom-local-coding-harnessradar:agenticos-self-hosted-governanceradar:agentgate-signed-agent-receiptsradar:anthropic-computer-use-apis-ga

queries asked of Scott's wikis

agent-harness skill architecture and skill-as-guardrail pattern
open-weight model sovereignty and local inference harnesses
agent memory and agent-maintained wiki patterns for skill registries
production guardrails: pre-action enforcement vs post-hoc audit
skillHarness / SkillHarness naming collision with Zhejiang University work
community skill distribution and adoption metrics for agent frameworks

Measured heat

now 0 pts/hpeak 1 pts/hcomments 0/hpeers p16momentum: steady2 platformsage 99h

points/hour across evidence · reading as of 2026-10-12 02:59:37.977291+11:00 · deterministic, not a model opinion

How the heat travelled

10-07 13:00	⭐ origin echo-reconstructed	Accountability.md – a skill that holds agents accountable for actions; released as a skill for agent harnesses. kenm47 on github (echo) · attributed from hn.story.50011080	—
10-08 19:48	first on hacker news · published · +30.8h	Show HN: Accountability.md – a skill that holds agents accountable for actions hank2000	—
10-08 19:48	amplified on hacker news 👑	hn.story.50011080 hank2000	peak 2 · 0 comments · 98% of case engagement
10-08 22:33	our radar first saw it · +33.5h	discovery anchor: hn.story.50011080	—

pace: p24 vs 1247 stories at the 96h mark (now 99h old) — ahead of aafp-commons-signed-agent-notebook (2.0x), behind agentgate-signed-agent-receipts (0.7x)

Evidence (2) — ⭐ canonical anchor

source	object	author	score	comments
🟧 hn	Show HN: Accountability.md – a skill that holds agents accountable for actions	hank2000	2	0
🟧 echo.github ⭐	Accountability.md – a skill that holds agents accountable for actions; released as a skill for agent harnesses.	kenm47	—	—

Interpretation history

2026-10-09T01:56:28Z
grounded: converges/high — A community-released accountability skill for agent harnesses directly implements Scott's load-bearing architectural position: governance must live in the execu
2026-10-09T01:43:09Z
case created — Show HN release of an agent-accountability skill directly targets a gap in deployed agent guardrails.

Decision trace

10-09 13:41feedback_briefingScott vote via UI
10-09 13:26attention_communicatedkenm47 released Accountability.md as a portable skill for agent harnesses — a deterministic, pre-action enforcement module that embodies the 'skill-as-guardrail' pattern Scott's framewo
10-09 13:26attention_routeFirst concrete community implementation of Scott's load-bearing architectural position on governance-as-architecture. Complements the Codex Instant Interrupts PR and ECC OS layer (also in this pa
10-09 13:18attention_routeFirst concrete community implementation of Scott's load-bearing architectural position on governance-as-architecture. Complements the Codex Instant Interrupts PR (also in this package) showing bo
10-09 13:11attention_candidatecreate
10-09 12:56groundA community-released accountability skill for agent harnesses directly implements Scott's load-bearing architectural position: governance must live in the execution path as deterministic, pre-act
10-09 12:43createShow HN release of an agent-accountability skill directly targets a gap in deployed agent guardrails.