AGENT SKILL

atlas-esl-hop

ESL lifecycle hop — the DISCOVER hop. When a scout mission's findings surface change-worthy signal (a defect, a spec/impl drift, or a genuine gap) in an ESL-enabled project (.spectra/ present), ATLAS proposes opening an ESL change at 'proposed' by framing the scout-report + its ECL envelope it already emits as the proposal artifact — it never calls mcp__tonberry__* itself and never implements (refusal boundary intact). SPECTRA's own esl-hop owns the actual right_size/propose/specify tonberry calls on receipt, over the existing atlas-to-spectra handoff edge. Absent .spectra/ or absent change-worthy findings → hand off the scout-report normally (ESL opt-in).

License 许可
Apache-2.0
所属仓库
Rynaro/magicite
最近更新
2026-10-14
Stars
★ 2
Forks
0

01 前言 Front Matter

name
atlas-esl-hop
metadata
{ "phase": "S", "methodology": "ATLAS" }
description
ESL lifecycle hop — the DISCOVER hop. When a scout mission's findings surface change-worthy signal (a defect, a spec/impl drift, or a genuine gap) in an ESL-enabled project (.spectra/ present), ATLAS proposes opening an ESL change at 'proposed' by framing the scout-report + its ECL envelope it already emits as the proposal artifact — it never calls mcp__tonberry__* itself and never implements (refusal boundary intact). SPECTRA's own esl-hop owns the actual right_size/propose/specify tonberry calls on receipt, over the existing atlas-to-spectra handoff edge. Absent .spectra/ or absent change-worthy findings → hand off the scout-report normally (ESL opt-in).

03 SKILL.md 正文

ATLAS — ESL Lifecycle Hop (the DISCOVER hop)

When to use

Load at Phase S (Synthesize), after the fold is validated and before scout-report.md is finalized, when both hold:

  • Change-worthy signal. ≥1 finding in findings.md asserts something is broken, drifted, or missing relative to a spec/expectation the mission surfaced — not merely descriptive topology. See "What counts as change-worthy" below.
  • ESL-enabled project. A .spectra/ directory exists at the consumer project root — checked with a single read-only list_dir probe, never an MCP call. ATLAS has no mcp__tonberry__* tool surface and never acquires one.

If either condition is false, skip this skill silently and emit scout-report.md the normal way (skills/synthesize/SKILL.md). ESL is opt-in; ATLAS is EIIS-standalone-conformant and works without it.

What counts as change-worthy

A finding is change-worthy when its claim is one of:

  • a defect — behavior contradicts a spec, contract, or documented invariant;
  • a drift — an implementation and its governing spec/doc have diverged;
  • a gap — a required capability, test, or safeguard is absent.

Purely descriptive findings ("X calls Y at path:line") are not change-worthy on their own. When uncertain, the mission's own GAPS/RISKS framing (scout-report §5) is the signal — anything already flagged there as needing a decision or a fix is change-worthy by definition.

Your hop — DISCOVER (proposed), never MAKE

You are the furthest-upstream Eidolon in the ESL chain. You do not call mcp__tonberry__* — right-sizing, opening the change.json record, and running S→P→E→C→T→R→A belong to SPECTRA's own hop (skills/esl-hop/SKILL.md in the SPECTRA repo), triggered when it receives your hand-off. Your job stops at naming the change-worthy finding and handing it off — the refusal boundary (P0 rule 1: read-only, no edit/write/commit/deploy/migrate/refactor/fix) stays fully intact: you never create, edit, or transition ESL state yourself.

  1. flag the proposal in §4. In scout-report.md §4 "Recommended next actions", tag every change-worthy R-N item [→ SPECTRA] with an explicit [ESL-PROPOSAL] marker naming the FINDING-XXX/GAP-XXX it rests on, and state plainly that the item proposes opening an ESL change at proposed. This is the same artifact Phase S already emits — no new file, no new schema.
  2. frame the envelope objective. In scout-report.envelope.json, set performative: "PROPOSE" (the existing default for ATLAS→SPECTRA) and word objective to name the ESL proposal intent, e.g. "Deliver scout report for mission <id>, proposing an ESL change at 'proposed' for FINDING-003/FINDING-007.". The ise block is unchanged from the standard scout-report emission (skills/synthesize/SKILL.md "Envelope sidecar"): ise.assertion_grade: "self-attested" and ise.receiver_authorization: {auto_route: true, auto_merge: false, auto_deploy: false} — SPECTRA MAY auto-route the finding into its own ESL intake without an operator confirm, but MUST NOT auto-merge or auto-deploy anything off an ATLAS-originated discovery.
  3. hand off via the existing edge. Deliver scout-report.md + scout-report.envelope.json to SPECTRA exactly as skills/synthesize/SKILL.md already prescribes (contracts/atlas-to-spectra.yaml, artifact.kind: scout-report, unchanged). SPECTRA's spectra-esl-hop skill (when .spectra/ and tonberry are present on its side) takes it from there: right_size → propose → specify.

Invariants

  • Refusal boundary intact. ATLAS never calls a tonberry write verb (propose, transition, archive, verify) and never touches .spectra/changes/. This hop is discovery framing on an artifact ATLAS already emits, not a new write surface.
  • SPECTRA owns proposed. The actual change.json at proposed status is created by SPECTRA's hop, never ATLAS's. ATLAS's contribution is evidence carried across the existing roster edge (contracts/atlas-to-spectra.yaml), not a tonberry-recognized object.
  • No new artifact kind, no new contract. The proposal rides inside the standard scout-report artifact/envelope pair — inventing a distinct ECL artifact.kind here would require a contract change outside ATLAS's repo; this hop deliberately does not do that.
  • Evidence-anchored, same as every ATLAS claim. Every [ESL-PROPOSAL] item cites a FINDING-XXX/GAP-XXX + its path:line anchor (I-7). No new claims are introduced at this hop.
  • Graceful skip. No .spectra/ directory, or no change-worthy finding → skip silently, emit scout-report.md normally. ESL is opt-in.

ATLAS — ESL Lifecycle Hop (the DISCOVER hop; upstream of proposed, never a maker)

04 同仓库更多 Skills

Rynaro
Rynaro/magicite

wine-dxvk-cache-clear

Clear a stale DXVK shader cache that is causing graphical corruption in a Wine or Proton game. Use when: a game shows corrupted textures or shader flicker that started after a driver or DXVK version change.

★ 2
Rynaro
Rynaro/magicite

steam-download-region-fix

Fix a Steam client stuck at very low download speed by changing its download region. Use when: Steam downloads are far below the local internet speed for every game, not just one.

★ 2
Rynaro
Rynaro/magicite

proton-battleye-eac-toggle

Toggle BattlEye or Easy Anti-Cheat runtime support for a Steam Proton game. Use when: a multiplayer game refuses to launch with an anti-cheat error under Proton.

★ 2
Rynaro
Rynaro/magicite

vivi-verify-incoming

Load when reading any upstream artefact handed off to Vivi that carries a sibling .envelope.json. BLOCKING per ECL §6.2.2: the orchestrator MUST have verified the envelope's SHA-256 (eidolons verify-envelope --block / eidolons run --verify) and recorded a verify_pass before dispatch. If no verify_pass exists for the message_id, or a verify_fail is present, REFUSE to process the payload and hand back to the orchestrator. Symmetric receiver gate — every Eidolon enforces it identically.

★ 2
Rynaro
Rynaro/magicite

vivi-parallel-tracks

Load ONLY under TRANCE authorization when the Plan phase yields N independent, non-overlapping implementation tracks (disjoint file sets). Operationalizes the TRANCE G4 form — parallel multi-track implementation in mandatory git-worktree isolation, max 5 clean-context tracks, a per-track verifier cascade, a non-fungible per-track ≤3 reflection budget, explicit stop conditions, and a single-threaded merge/aggregation step. Absent TRANCE gating, Vivi runs exactly as today: single-track A→P→I→V→Δ/R.

★ 2
Rynaro
Rynaro/magicite

vivi-methodology

Full Vivi cycle reference — A(nalyze) → P(lan) → I(mplement) → V(erify) → Δ(Delta)/R(eflect). Use when the task is a non-trivial feature implementation in a brownfield codebase and you need the complete methodology reference: complexity routing, evidence-grounded planning rules, test-anchoring requirements, and failure-escalation thresholds.

★ 2