Pypi Release
alchemiststudiosDOTai/tunacode
This skill should be used when releasing tunacode-cli to PyPI.
Operating conventions shared by every role - delegating, committing, asking, running tests, model tiers, and artifact ownership.
Install Agent Runtime
$ npx skills add oocx/tfplan2md --skill agent-runtime -a claude-codeProject install by default; add -g for ~/.claude/skills/.
$ gh skill install oocx/tfplan2md agent-runtime --agent claude-codeProject scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).
$ git clone --depth 1 https://github.com/oocx/tfplan2md.git skills-src && mkdir -p .claude/skills && cp -r skills-src/.agents/skills/agent-runtime .claude/skills/agent-runtime && rm -rf skills-srcUse ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.
Claude Code skills documentation · loads skills from .claude/skills/
Install the "agent-runtime" agent skill from https://github.com/oocx/tfplan2md/tree/main/.agents/skills/agent-runtime into .claude/skills/agent-runtime/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "agent-runtime", then confirm the skill loads.Claude Code copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$skill-installer install https://github.com/oocx/tfplan2md/tree/main/.agents/skills/agent-runtimeType this inside Codex. $skill-installer <name> installs a curated skill from openai/skills. The installer writes to $CODEX_HOME/skills (default ~/.codex/skills). Restart Codex if the skill does not show up.
$ npx skills add oocx/tfplan2md --skill agent-runtime -a codexProject install goes to .agents/skills/; add -g for ~/.codex/skills/.
$ gh skill install oocx/tfplan2md agent-runtime --agent codexProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/oocx/tfplan2md.git skills-src && mkdir -p .agents/skills && cp -r skills-src/.agents/skills/agent-runtime .agents/skills/agent-runtime && rm -rf skills-srcUse ~/.agents/skills/ instead of .agents/skills for a personal install.
Codex skills documentation · loads skills from .agents/skills/
Install the "agent-runtime" agent skill from https://github.com/oocx/tfplan2md/tree/main/.agents/skills/agent-runtime into .agents/skills/agent-runtime/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "agent-runtime", then confirm the skill loads.Codex copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ npx skills add oocx/tfplan2md --skill agent-runtime -a cursorProject install goes to .agents/skills/; add -g for ~/.cursor/skills/.
$ gh skill install oocx/tfplan2md agent-runtime --agent cursorProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/oocx/tfplan2md.git skills-src && mkdir -p .cursor/skills && cp -r skills-src/.agents/skills/agent-runtime .cursor/skills/agent-runtime && rm -rf skills-srcUse ~/.cursor/skills/ instead of .cursor/skills for a personal install.
Cursor skills documentation · loads skills from .cursor/skills/, .agents/skills/, .claude/skills/, .codex/skills/
Install the "agent-runtime" agent skill from https://github.com/oocx/tfplan2md/tree/main/.agents/skills/agent-runtime into .cursor/skills/agent-runtime/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "agent-runtime", then confirm the skill loads.Cursor copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ gemini skills install https://github.com/oocx/tfplan2md.git --path .agents/skills/agent-runtime--scope user (default) or --scope workspace; --path is the subfolder of the repo that holds the skill; --consent skips the security confirmation prompt.
$ npx skills add oocx/tfplan2md --skill agent-runtime -a gemini-cliProject install goes to .agents/skills/; add -g for ~/.gemini/skills/.
$ gh skill install oocx/tfplan2md agent-runtime --agent gemini-cliProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/oocx/tfplan2md.git skills-src && mkdir -p .gemini/skills && cp -r skills-src/.agents/skills/agent-runtime .gemini/skills/agent-runtime && rm -rf skills-srcUse ~/.gemini/skills/ instead of .gemini/skills for a personal install, then run /skills reload.
Gemini CLI skills documentation · loads skills from .gemini/skills/, .agents/skills/
Install the "agent-runtime" agent skill from https://github.com/oocx/tfplan2md/tree/main/.agents/skills/agent-runtime into .gemini/skills/agent-runtime/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "agent-runtime", then confirm the skill loads.Gemini CLI copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ gh skill install oocx/tfplan2md agent-runtimeInstalls for Copilot at project scope by default; add --scope user for a personal install. Preview a skill first with gh skill preview. Needs GitHub CLI 2.90.0 or later (public preview).
$ npx skills add oocx/tfplan2md --skill agent-runtime -a github-copilotProject install goes to .agents/skills/; add -g for ~/.copilot/skills/.
$ git clone --depth 1 https://github.com/oocx/tfplan2md.git skills-src && mkdir -p .github/skills && cp -r skills-src/.agents/skills/agent-runtime .github/skills/agent-runtime && rm -rf skills-srcUse ~/.copilot/skills/ instead of .github/skills for a personal install. Commit .github/skills so cloud agent and code review can use it.
GitHub Copilot skills documentation · loads skills from .github/skills/, .claude/skills/, .agents/skills/
Install the "agent-runtime" agent skill from https://github.com/oocx/tfplan2md/tree/main/.agents/skills/agent-runtime into .github/skills/agent-runtime/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "agent-runtime", then confirm the skill loads.GitHub Copilot copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ npx skills add oocx/tfplan2md --skill agent-runtime -a opencodeOpenCode documents no install command of its own. Project install goes to .agents/skills/; add -g for ~/.config/opencode/skills/.
$ gh skill install oocx/tfplan2md agent-runtime --agent opencodeProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/oocx/tfplan2md.git skills-src && mkdir -p .opencode/skills && cp -r skills-src/.agents/skills/agent-runtime .opencode/skills/agent-runtime && rm -rf skills-srcUse ~/.config/opencode/skills/ instead of .opencode/skills for a personal install.
OpenCode skills documentation · loads skills from .opencode/skills/, .claude/skills/, .agents/skills/
Install the "agent-runtime" agent skill from https://github.com/oocx/tfplan2md/tree/main/.agents/skills/agent-runtime into .opencode/skills/agent-runtime/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "agent-runtime", then confirm the skill loads.OpenCode copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
Facts
- Skill name
agent-runtime- GitHub stars
- 174
- Token cost
- ~1.4k tokens
- SKILL.md length
- 722 words
- Files
- 1
- Skills in repo
- 28
- Repo updated
- First seen
- Licence
- MIT
At a glance
Operating conventions shared by every role - delegating, committing, asking, running tests, model tiers, and artifact ownership.
- Development work in your project
- SKILL.md covers Running as a role, Model tiers, Committing and Asking, plus 4 more sections
- Calls git and dotnet
What it does
Agent Runtime is an agent skill from oocx/tfplan2md. Operating conventions shared by every role - delegating, committing, asking, running tests, model tiers, and artifact ownership. Load this before acting as any role.
Its SKILL.md is about 1.4k tokens, which your agent loads only when the skill is triggered. It is a single SKILL.md file with no bundled scripts.
It sits in Development. It works with OpenAI. The repository describes itself as: Convert terraform plans (json) into human readable markdown for easier review of changes in pull requests. The licence is MIT.
When your agent uses it
- Development work in your project
Example prompts
- “/agent-runtime”
What it can do on your machine
Read from SKILL.md and the folder at commit 88d5240. It shows what the files ask for, not the result of running them.
Tool permissions
Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.
From allowed-tools in the SKILL.md frontmatter.
Runs code
Shell commands in SKILL.md call:
gitdotnet
From the folder's file list and the shell code blocks in SKILL.md.
Network
No URLs in SKILL.md. Its commands use git, which can reach the network depending on how they are called.
From URLs in SKILL.md, links to its own repository left out.
Credentials
Names no API keys, tokens, secrets or passwords.
From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.
Context cost
Agent Runtime loads about 1.4k tokens when it runs. Until then it costs about 45 tokens; SKILL.md has 722 words of instructions outside code blocks.
Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.
Safety
Auto-check passed
The automated check found no risky patterns in SKILL.md.
Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.
SKILL.md
The full file from oocx/tfplan2md at commit 88d5240, republished under its MIT licence (© oocx). 722 words, ~1,392 tokens.
Download SKILL.mdSave it as
.claude/skills/agent-runtime/SKILL.md (or your agent's skills folder).- name
- agent-runtime
- description
- Operating conventions shared by every role - delegating, committing, asking, running tests, model tiers, and artifact ownership. Load this before acting as any role.
Agent Runtime
Conventions every role obeys. Role files under .agents/roles/ state only what is
specific to their role and defer everything here. AGENTS.md
carries the project-wide coding, commit and terminal rules; this skill covers how a
role operates.
This is also the only file that needs to change to support a second harness.
Running as a role
Prefer spawning the role as a subagent over adopting it in-session. The isolated context is the main reason this workflow is affordable: a stage's file reads, test output and search results never enter the parent thread.
Adopt a role in-session only when you need its output in your working context to continue immediately.
Read the role file before acting as one. Do not improvise a role's boundaries — they exist to keep artifact ownership clean, and the work-protocol gate checks that the right roles ran.
Model tiers
.agents/tiers.json is the single source of truth. Role files declare tier:, never
a model name.
| Tier | Claude | Codex |
|---|---|---|
| deep | opus | gpt-5.6-sol |
| standard | sonnet | gpt-5.6-terra |
| cheap | haiku | gpt-5.6-luna |
Codex needs the full model slug. Bare sol / terra / luna are rejected by the API
("not supported when using Codex with a ChatGPT account").
Escalation: on rework attempt 2 or later, run the role one tier deeper. Cheap models get first crack; expensive ones only see the cases that failed once.
Delegate to the cheap tier for work whose output you do not need verbatim: codebase search, "does this file exist", build and test runs where only pass/fail matters. Do not delegate single file reads or one grep — the spawn costs more than the call.
Committing
Commit with git directly. Conventional Commits, types per AGENTS.md.
- Commit before handing off. The next role must see complete work.
- Amend, don't stack fixups. To correct work you just committed, use
git commit --amend. Never create a "fix the previous commit" commit. - Commit only files your role owns (see below).
Asking
The workflow runs unattended. There are exactly three gates — specification approval, a genuinely contested architecture choice, and UAT for user-visible output changes. They are defined in docs/workflow.md.
Away from a gate, do not block. Record the question and the assumption you are
proceeding on in state.json → open_questions, then continue:
scripts/wp-append.sh --question "Should X do Y?" --assumed "Yes, matching Z"The accumulated list surfaces at the next gate and in the PR description. This is a deliberate inversion of the old "ask one question at a time and wait" rule, which made unattended runs impossible.
At a gate, ask one question at a time and wait. If you offer alternatives, give pros and cons and a recommendation with its reason.
When genuinely blocked: say you are blocked in one sentence, summarise what is done, state what remains, and ask one question.
Show full SKILL.md (275 more words)Show less
Running tests
Use the run-dotnet-tests skill. Never call dotnet test directly — .NET 10 has two
test runners and a direct call from the repo root fails with MSB1001.
Test naming: MethodName_Scenario_ExpectedResult. TUnit is the only test framework in
this project.
Work protocol
Every role appends an entry to <work-item>/work-protocol.md after completing its work
and before handing off:
scripts/wp-append.sh --role "Developer" \
--summary "..." --artifacts "..." --problems "None"Record problems honestly — the Retrospective reads them, and a problem that was never
logged cannot be fixed. scripts/workflow-gate.sh blocks release when a required role
has no entry.
Artifact ownership
Each artifact has exactly one owning role. Do not edit another role's artifacts; hand back instead.
| Artifact | Owner |
|---|---|
src/, src/tests/ | Developer |
docs/, README.md (global docs) | Technical Writer |
docs/workflow.md, AGENTS.md | Workflow Engineer |
specification.md | Requirements Engineer |
analysis.md | Issue Analyst |
architecture.md, docs/adr-*.md | Architect |
test-plan.md, uat-test-plan.md | Quality Engineer |
tasks.md | Task Planner |
code-review.md | Code Reviewer |
uat-plan.json, uat-plan.md | Developer (to the Quality Engineer's specification) |
uat-report.md | UAT Tester |
| UAT PRs and their comments | UAT Tester |
release-notes.md, the PR, the release | Release Manager |
retrospective.md | Retrospective |
.agents/, AGENTS.md, scripts/ workflow tooling | Workflow Engineer |
website/src/ | Web Designer |
work-protocol.md is append-only, and only through scripts/wp-append.sh.
state.json belongs to the driver. Roles never write stage, status or gates.*
directly — a role that could write a gate field could clear a rejection the Maintainer
had already made. The one field a role sets is arch_contested, and only the Architect
sets it.
Reporting
End a turn with what changed, what is next, and what you need — or "Nothing" if unblocked. Keep it short; the work-protocol entry is the durable record, not the chat.
When handing off, state what you completed, which files you touched, and the specific next action for the receiving role.
© oocx, MIT. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file
Files
Just SKILL.md in .agents/skills/agent-runtime of oocx/tfplan2md.
Open the folder on GitHubat commit 88d5240
Compare with similar skills
Agent Runtime next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.
| Skill | Stars | Used in | Tokens | Auto-check | Licence | Repo updated |
|---|---|---|---|---|---|---|
| Agent Runtime this skilloocx/tfplan2md | 174 | — | ~1.4k | Automated safety check: Pass | MIT | |
| Pypi ReleasealchemiststudiosDOTai/tunacode | 125 | — | ~2.2k | Automated safety check: Pass | MIT | |
| Goinference-gateway/inference-gateway | 214 | — | ~2.4k | Automated safety check: Pass | Apache-2.0 | |
| Go Concurrencyinference-gateway/inference-gateway | 214 | — | ~2.3k | Automated safety check: Pass | Apache-2.0 | |
| Juror ReviewJuror-AI/juror | 120 | — | ~576 | Automated safety check: Pass | MIT | |
| Initializefrenzymath/Danus | 475 | — | ~1.4k | Automated safety check: Pass | Apache-2.0 |
Similar skills
Go
inference-gateway/inference-gateway
Idiomatic Go - package and interface design, error wrapping, table-driven tests, generics, the modern standard library (slices/maps/cmp/errors.Join), current syntax, and logging discipline.
Go Concurrency
inference-gateway/inference-gateway
Idiomatic Go concurrency - sync primitives, channel semantics, the select statement, and the standard channel patterns (cancellation/done-channel, fan-out/fan-in, pipeline, or-done, context…
Juror Review
Juror-AI/juror
Inspect Juror Cloud PR findings and, only after an explicit confirmation, start or rerun a hosted Juror review.
Initialize
frenzymath/Danus
First-run setup interview for a Danus deployment. An agent skill from frenzymath/Danus.
Genie API Service Docs
qualcomm/qai-appbuilder
GenieAPIService technical documentation retrieval. An agent skill from qualcomm/qai-appbuilder.
More from oocx/tfplan2md
All 28 skills in this repoCreate Agent Skill
oocx/tfplan2md
Create a new Agent Skill following project standards and templates.
Detect Diagram Crossings
oocx/tfplan2md
Detect and analyze edge crossings and overlaps in SVG workflow diagrams using geometric intersection algorithms and visual analysis.
Generate Release Screenshots
oocx/tfplan2md
Generate PNG screenshots for release notes using the repository's HtmlRenderer and ScreenshotGenerator tools.
Git Rebase Main
oocx/tfplan2md
Safely rebase the current feature branch on top of the latest origin/main.
Next Issue Number
oocx/tfplan2md
Determine the next available issue number across all change types (feature, fix, workflow, website) by checking both local docs and remote branches, then reserve it by pushing an empty branch.
Run Uat
oocx/tfplan2md
Run User Acceptance Testing by creating a PR with rendered markdown on GitHub or Azure DevOps.
Related
Works with
Categories
Questions about Agent Runtime
What does Agent Runtime do?
Operating conventions shared by every role - delegating, committing, asking, running tests, model tiers, and artifact ownership. Agent Runtime is an agent skill from oocx/tfplan2md. Operating conventions shared by every role - delegating, committing, asking, running tests, model tiers, and artifact ownership.
When should I use Agent Runtime?
Agent Runtime fits situations like: development work in your project.
How do I install Agent Runtime in Claude Code?
Run `npx skills add oocx/tfplan2md --skill agent-runtime -a claude-code`. Or copy the skill folder (.agents/skills/agent-runtime in oocx/tfplan2md) into .claude/skills/agent-runtime in your project. Claude Code loads it when a task matches its description.
How do I install Agent Runtime in Codex?
Run `npx skills add oocx/tfplan2md --skill agent-runtime -a codex`. Or copy the skill folder (.agents/skills/agent-runtime in oocx/tfplan2md) into .agents/skills/agent-runtime in your project. Codex loads it when a task matches its description.
Can I use Agent Runtime in Cursor, Gemini CLI or GitHub Copilot?
Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add oocx/tfplan2md --skill agent-runtime -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/agent-runtime, .gemini/skills/agent-runtime, .github/skills/agent-runtime and .opencode/skills/agent-runtime in your project.
What does Agent Runtime need to run?
Going by SKILL.md and its folder, Agent Runtime needs the command-line tools its instructions call (git and dotnet).
Does Agent Runtime access the network?
SKILL.md contains no URLs. Its commands use git, which can reach the network depending on how they are called. This is read from the text; nothing was executed.
Is Agent Runtime safe to install?
Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.
What licence does Agent Runtime use?
Agent Runtime is published under the MIT licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.
How many tokens does Agent Runtime use?
About 1.4k tokens (SKILL.md is roughly 5.6k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full.
What are the alternatives to Agent Runtime?
Skills that share tags, products or a category with Agent Runtime: Pypi Release (alchemiststudiosDOTai/tunacode, 125 stars), Go (inference-gateway/inference-gateway, 214 stars), Go Concurrency (inference-gateway/inference-gateway, 214 stars) and Juror Review (Juror-AI/juror, 120 stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.
Who maintains Agent Runtime?
oocx (a GitHub user) maintains it in oocx/tfplan2md, which has 174 GitHub stars. The repository holds 28 skills in this directory. The repository was last updated on October 7, 2026.
Source: oocx/tfplan2md on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.