Skip to content
Skills Library
Programming · Building & coding

deep-implement

/deep-implement

Use when implementing dependency-ready deep-work slices with their sealed strict or outcome verification basis.

Repository
deep-work ↗
Licence
MIT
Last repository change
14.10.2026

Assessment and security

81/ 100Class: good
Format25 / 25
Frontmatter with name and description, valid name, description up to 1,024 characters, name matching the folder.
Discoverability25 / 25
Description explains when to use the skill, has a reasonable length and uses third person.
Content16 / 35
At least 150 words and at most 500 lines, headings, step lists, examples or code.
Hygiene15 / 15
No placeholders such as TODO, no Windows paths, with a licence.
Riskno signals detected in the scan
Files in this package
5
Scripts
0
HTML comments
0
URLs in SKILL.md
0
Deprecation

No signals detected. Repository changed on 14.10.2026, without references to retired models or APIs.

Licence

MIT, open: use and modify while retaining author attribution.

What it does and when to use it

Use when implementing dependency-ready deep-work slices with their sealed strict or outcome verification basis.

Original author description

Use when implementing dependency-ready deep-work slices with their sealed strict or outcome verification basis.

How it works

SKILL.md has no level-two headings · 19 lines, 328 words

SKILL.md preview
---
name: deep-implement
description: "Use when implementing dependency-ready deep-work slices with their sealed strict or outcome verification basis."
user-invocable: true
---

# Implement and verify each sealed outcome

Resolve the plugin root to a literal absolute path. Before reading or running any plugin file, resolve its real path and require containment in that root. All `${CLAUDE_PLUGIN_ROOT}` paths below mean that verified root.

Resolve `--session=ID` from `$ARGUMENTS`; otherwise run `node "${CLAUDE_PLUGIN_ROOT}/scripts/deep-work-runtime.js" session context`. Read the returned state path and its `work_dir`; never guess a work directory or change the active pointer to another session. Validate with `session authority validate --state "$STATE_FILE"`. Use the runtime's phase-aware result: a legitimate pre-plan session does not require a plan that has not been created.

Read `${CLAUDE_PLUGIN_ROOT}/skills/shared/references/runtime-execution-spine.md`. The runtime owns phase, approval, write, review and receipt mutations. Run `node "${CLAUDE_PLUGIN_ROOT}/scripts/deep-work-runtime.js" --help` for exact current route arguments; consume returned IDs/digests rather than inventing them.

Read `$WORK_DIR/plan.md`, its runtime projection and `${CLAUDE_PLUGIN_ROOT}/skills/shared/references/implementation-guide.md`. Preserve the requested current model and effort; explicit available delegation follows `${CLAUDE_PLUGIN_ROOT}/skills/deep-implement/references/execution-mode.md`.

Select an incomplete dependency-ready slice through `slice activate`. A session has one active slice/write window. Open the appropriate scope with `implement write begin`, edit only those files, and immediately call `implement write accept` with the returned operation/pre-manifest identity. Failed acceptance requires recovery/replan; never mark the edit accepted yourself.

For `strict-tdd-v2`, use actual failing-test writes → `verification run-v2` → `verification red-transition` → `verification proof-publish` before production writes. Then run actual GREEN, applicable sensors/refactor checks and `implement slice complete-v2`. Unsupported policy/Node/runner is an explicit capability result, not permission to fake RED.

For `outcome-v1`, do not run the TDD transition path. After accepted source writes (or `outcome source observe` for already-correct source), use `verification outcome-explain` and explicit `verification outcome-run --prepared-digest ...`. Authenticate positive/control results, publish the required per-ID oracle review and call `implement outcome-complete`. Both views use the final oracle/dependencies; accepted production files are never reverted for a counterexample.

Completion is the authenticated internal receipt plus its public M3 publication. Recover missing publication instead of repeating code. Continue dependency-ready slices, then the Test phase. Report actual completed outcomes and unresolved verification, not instructions or synthetic counts.
Files in this package (5)
  • SKILL.md 3 KB
  • references/execution-mode.md 1 KB
  • references/rollback-protocol.md 1 KB
  • references/tdd-red-flags.md 2 KB
  • references/team-path.md 1 KB

Versions

RepositoryScoreLicence
This page · sungmin-cho/deep-work81MIT
piercelamb/deep-implement100MIT

Database snapshot: 13.11.2026 · package version 2ee2bfa34a60 (SHA-256 digest of the entire package)

Downloads in the last 30 days: 0 · ranking

Report a problem with this skill: harmful content, incorrect descriptions or an author’s removal request.

Similar skills

  • 81deep-test
    Use when validating a deep-work goal, consuming completion evidence or retrying failed slices.
  • 81deep-work-orchestrator
    Use when starting or continuing a deep-work task with durable goal, verification and completion evidence.
  • 100flutter-navigation
    Implement navigation patterns with go_router, deep linking, and named routes in Flutter. Use when building navigation, deep linking, or routing.
  • 85implement-bd
    Implement Beads issues and record verification and remaining work. Use when asked to work on a tracked issue or pick up ready Beads work.
  • 82implement-work
    Use to implement one PR-sized task.
  • 90mobile-deep-linking
    Use this skill when the user says 'deep linking', 'universal link', 'app link', 'URL scheme', 'deferred deep link', 'routing deep link', 'link handler', 'deep navigation'. Implement deep linking across iOS and Android using universal links, custom URL schemes, and deferred deep…

Feedback and reports

Your message goes to Bartosz Sokoliński. An email address is needed only for a reply. See the privacy policy.