oracle-first
Verification-economics method for AI-assisted work.
Skill by marcelocantos
Install
Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.
npx agentmarket install skill/marcelocantos-oracle-first
Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.
Get the weekly diff
The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.
About
Reviews
Write a reviewNo reviews yet. Installed it? Reviews come from verified installers only. How reviews work
Permissions
What this skill can do and what it asks for, from its SKILL.md. Checked Oct 25, 2026.
- Tools
- No tools found
- Secrets
- None asked for
- Install scripts
- None found
- Runs as
- Python package
Network hosts it talks to 1
codeload.github.com· found in install command
Files it touches
~/.claude/skills/oracle-first
What's inside
- Skill name
oracle-first- Description
- Verification-economics method for AI-assisted work. Use when porting or migrating legacy code, replicating an existing system's behaviour ("match the old app exactly"), doing visual/physics/feel parity work (especially spatial/geometry), authoring the correctness spec for new code (property tests, invariants, TLA+), designing acceptance criteria for a target, planning verification for a codebase analysis, or when repeated tweak-and-check against human judgment isn't converging.
- License
- unknown
- Instructions
- 406 lines in SKILL.md · view on GitHub
Before installing
- Ships executable scripts
Files (6)
- SKILL.md
- doctrine.md
- honesty-ratchet.md
- instrument-error.md
- README.md · ratchet/
- ratchet.py · ratchet/
The license isn't open source, so the instructions aren't shown here. Read SKILL.md on GitHub
More from marcelocantos/skills
- 0.559score
stability
marcelocantos
Update one repo's STABILITY.md onto the current 1.0 approach, or, once it is current, review that contract and tell the owner what their `## 1.0 judgment` has…
SkillAFreshness: AClaude Code +9 - 0.553score
progress-report
marcelocantos
Generate and publish a weekly progress report from git activity across all repos.
SkillAFreshness: AClaude Code +9 - 0.540score
release
marcelocantos
Publish a release — version, release notes, CI, Homebrew tap, tag, and GitHub release.
SkillAFreshness: AClaude Code +9 - 0.524score
entropy-audit
marcelocantos
Audit a whole repository or substantial code change for structural entropy: unclear top-level architecture, boundary violations, dependency cycles, redundant…
SkillBFreshness: BClaude Code +9 - 0.519score
expenses
marcelocantos
Find Gmail receipts/invoices for a date range, let the user select which are work-related, and forward only confirmed items to the CBA work email.
SkillBFreshness: BClaude Code +9 - 0.519score
build-perf-audit
marcelocantos
Audit a build system for performance issues — profile the build, identify hotspots, match against known anti-patterns, and propose (or apply) fixes with…
SkillBFreshness: BClaude Code +9 - 0.516score
retro
marcelocantos
Mine the last week of session transcripts for concrete improvements to the system itself — skills, tools, MCP servers, scripts, CLAUDE.md/AGENTS.md…
SkillBFreshness: BClaude Code +9 - 0.516score
remind
marcelocantos
Schedule a reminder that fires as a Slack DM at a future time.
SkillBFreshness: BClaude Code +9 - 0.516score
hygiene
marcelocantos
Declare and drift-validate a repo's steady-state hygiene posture (tests, security scans, code quality, governance, …) and aggregate coverage across the fleet.
SkillBFreshness: BClaude Code +9 - 0.516score
pr-audit
marcelocantos
Audit open PRs across all owned repos and recommend cleanup actions (close superseded, merge ready, poke contrib reviewers, fan-in synchronized rollouts).
SkillBFreshness: BClaude Code +9 - 0.513score
c
marcelocantos
Continue — restore compacted context from this session's chain after a /clear.
SkillBFreshness: BClaude Code +9 - 0.513score
docs
marcelocantos
End-to-end documentation sherpa — audit, plan, and write all project documentation.
SkillBFreshness: BClaude Code +9
Signals
- Stars
- 0
- Forks
- 0
- Open issues
- 0
- Last commit (this folder)
- Sep 12, 2026 (1 month ago)
From GitHub (fetched Oct 24, 2026).
Details
- License
- License unknown
- Repository
- github.com/marcelocantos/skills
- First seen
- Oct 24, 2026
- Last checked
- Oct 24, 2026
- Source
- GitHubgithub-graphql
More from marcelocantos
- 0.559score
stability
marcelocantos
Update one repo's STABILITY.md onto the current 1.0 approach, or, once it is current, review that contract and tell the owner what their `## 1.0 judgment` has…
SkillAFreshness: AClaude Code +9 - 0.553score
progress-report
marcelocantos
Generate and publish a weekly progress report from git activity across all repos.
SkillAFreshness: AClaude Code +9 - 0.540score
release
marcelocantos
Publish a release — version, release notes, CI, Homebrew tap, tag, and GitHub release.
SkillAFreshness: AClaude Code +9 - 0.524score
entropy-audit
marcelocantos
Audit a whole repository or substantial code change for structural entropy: unclear top-level architecture, boundary violations, dependency cycles, redundant…
SkillBFreshness: BClaude Code +9 - 0.519score
build-perf-audit
marcelocantos
Audit a build system for performance issues — profile the build, identify hotspots, match against known anti-patterns, and propose (or apply) fixes with…
SkillBFreshness: BClaude Code +9 - 0.519score
expenses
marcelocantos
Find Gmail receipts/invoices for a date range, let the user select which are work-related, and forward only confirmed items to the CBA work email.
SkillBFreshness: BClaude Code +9
Cite this page
oracle-first skill. Agent Mart. https://agentmart.dev/skills/marcelocantos-oracle-first (last updated 2026-10-24)