New: 98,000+ skills indexed from GitHub repos with 100+ stars · every skill safety-checkedBrowse top skillsTop skills
Esc to closeBrowse all skills
Submit a skillBrowse skills
Submit a skillBrowse skills

Agent skill

Recursive Mode

by try-works in try-works/recursive-mode

Repository workflow orchestration skill for staged implementation, locked artifacts, late-phase receipts, and durable memory maintenance.

Apache-2.0Auto-check passedDevelopment

Install Recursive Mode

skills CLI
$ npx skills add try-works/recursive-mode --skill recursive-mode -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI
$ gh skill install try-works/recursive-mode recursive-mode --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Manual copy
$ git clone --depth 1 https://github.com/try-works/recursive-mode.git skills-src && mkdir -p .claude/skills && cp -r skills-src/skills/recursive-mode .claude/skills/recursive-mode && rm -rf skills-src

Use ~/.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/

Facts

Skill name
recursive-mode
GitHub stars
136
Token cost
~3.4k tokens
SKILL.md length
1,384 words
Files
66 (incl. scripts, references)
Skills in repo
9
Repo updated
First seen
Licence
Apache-2.0

At a glance

Repository workflow orchestration skill for staged implementation, locked artifacts, late-phase receipts, and durable memory maintenance.

  • Works in 8 steps: Read /.recursive/RECURSIVE.md. → Read /.codex/AGENTS.md as the primary… → Read any other AGENTS.md copies only as… → …
  • Executing recursive-mode runs
  • SKILL.md covers Purpose, Bootstrap, Read Order and Operational Surface, plus 5 more sections
  • Runs PowerShell, Python and Shell scripts from its folder; calls npx, python and python3

What it does

Recursive Mode is an agent skill from try-works/recursive-mode. Repository workflow orchestration skill for staged implementation, locked artifacts, late-phase receipts, and durable memory maintenance. Use when executing recursive-mode runs, resuming a run, locking a phase, or verifying locks.

Its SKILL.md is about 3.4k tokens, which your agent loads only when the skill is triggered. The skill folder holds 69 other files, including scripts and reference files (for example `agents/openai.yaml`, `references/agents-block.md` and `references/artifact-template.md`).

It sits in Development. The repository describes itself as: Recursive workflow for agentic engineering. Like Factory Missions but properly recursive, free and open source. The licence is Apache-2.0.

When your agent uses it

  • Executing recursive-mode runs
  • Locking a phase
  • Verifying locks

Example prompts

  • “/recursive-mode”

Requirements

  • Python 3
  • Node.js
  • A Bash shell
  • PowerShell

Workflow steps

8 steps, taken from the first numbered list in SKILL.md.

  1. Read /.recursive/RECURSIVE.md.
  2. Read /.codex/AGENTS.md as the primary Codex bridge if present.
  3. Read any other AGENTS.md copies only as mirrored bridge guidance if the repo uses them.
  4. Read /.agent/PLANS.md as the primary Codex plans bridge if present.
  5. Read /.recursive/STATE.md.
  6. Read /.recursive/DECISIONS.md.
  7. Read /.recursive/memory/MEMORY.md.
  8. Load only the relevant memory docs under /.recursive/memory/.

What it can do on your machine

Read from SKILL.md and the folder at commit ff75bc7. 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

    Ships 11 files in scripts/ (PowerShell, Python and Shell, from the files we listed), which the agent can run.

    Shell commands in SKILL.md call:

    • npx
    • python
    • python3
    • bash
    • pwsh

    From the folder's file list and the shell code blocks in SKILL.md.

  • Network

    No URLs in SKILL.md. Its commands use npx, 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

Recursive Mode loads about 3.4k tokens when it runs, and up to ~51k if it reads all its reference files. Until then it costs about 61 tokens; SKILL.md has 1,384 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~61
When it runs · the whole SKILL.md, loaded when a task matches
~3.4k
With references · SKILL.md plus every file in references/, read only if the agent opens them
~51k

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); the scripts in this folder are not scanned.

SKILL.md

The full file from try-works/recursive-mode at commit ff75bc7, republished under its Apache-2.0 licence (© try-works). 1,384 words, ~3,381 tokens.

Download SKILL.mdSave it as .claude/skills/recursive-mode/SKILL.md (or your agent's skills folder). This skill also uses 65 other files; get the full folder from GitHub.
name
recursive-mode
description
Repository workflow orchestration skill for staged implementation, locked artifacts, late-phase receipts, and durable memory maintenance. Use when executing recursive-mode runs, resuming a run, locking a phase, or verifying locks.

recursive-mode

Purpose

Use this skill to run the repository workflow defined in /.recursive/RECURSIVE.md.

This file is the installable entrypoint for the skill. It is not the canonical workflow spec. For phase order, audit-loop rules, lock rules, addenda, late-phase receipts, state/decisions updates, and memory maintenance, follow /.recursive/RECURSIVE.md.

Bootstrap

Before doing recursive-mode work, ensure the repo scaffold exists:

  • /.recursive/RECURSIVE.md
  • /.recursive/STATE.md
  • /.recursive/DECISIONS.md
  • /.recursive/memory/MEMORY.md
  • /.recursive/memory/skills/SKILLS.md
  • /.recursive/memory/{domains,patterns,incidents,episodes,archive}/
  • /.recursive/memory/skills/{availability,usage,issues,patterns}/
  • /.recursive/run/
  • /.codex/AGENTS.md primary Codex bridge doc
  • /.agent/PLANS.md bridge doc
  • optional mirrored AGENTS.md docs if the repo already uses them

If any of those are missing or stale, the agent should bootstrap the repo automatically before continuing. Do not ask the user to run the bootstrap manually unless no supported Python or shell runtime is available in the environment. PowerShell is optional, not required.

Preferred bootstrap commands:

bash
python "<SKILL_DIR>/scripts/install-recursive-mode.py" --repo-root .
python3 "<SKILL_DIR>/scripts/install-recursive-mode.py" --repo-root .
bash "<SKILL_DIR>/scripts/install-recursive-mode.sh" --repo-root .
powershell -ExecutionPolicy Bypass -File "<SKILL_DIR>/scripts/install-recursive-mode.ps1" -RepoRoot .
pwsh -NoProfile -File "<SKILL_DIR>/scripts/install-recursive-mode.ps1" -RepoRoot .

Important boundary:

  • npx skills add ... installs the skill package into agent directories.
  • The target repository scaffold should then be created automatically on first recursive-mode use or via a wired session-start hook.
  • Python and Bash paths are first-class and should work on macOS and Linux without PowerShell.
  • Manual bootstrap commands are the fallback path, not the preferred normal UX.

Read Order

  1. Read /.recursive/RECURSIVE.md.
  2. Read /.codex/AGENTS.md as the primary Codex bridge if present.
  3. Read any other AGENTS.md copies only as mirrored bridge guidance if the repo uses them.
  4. Read /.agent/PLANS.md as the primary Codex plans bridge if present.
  5. Read /.recursive/STATE.md.
  6. Read /.recursive/DECISIONS.md.
  7. Read /.recursive/memory/MEMORY.md.
  8. Load only the relevant memory docs under /.recursive/memory/.

If any bridge doc conflicts with /.recursive/RECURSIVE.md, follow /.recursive/RECURSIVE.md.

Operational Surface

  • Canonical run root: /.recursive/run/<run-id>/
  • Canonical workflow spec: /.recursive/RECURSIVE.md
  • Canonical state: /.recursive/STATE.md
  • Canonical decisions ledger: /.recursive/DECISIONS.md
  • Canonical memory router: /.recursive/memory/MEMORY.md

Trigger Examples

  • Implement requirement '<run-id>'
  • Run Recursive Phase 2 for .recursive/run/<run-id>/
  • Resume requirement '<run-id>' after manual QA
  • Verify locks for .recursive/run/<run-id>/

Utility Scripts

  • scripts/install-recursive-mode.py
  • scripts/install-recursive-mode.ps1
  • scripts/install-recursive-mode.sh
  • scripts/recursive-init.py
  • scripts/recursive-init.ps1
  • scripts/recursive-status.py
  • scripts/recursive-status.ps1
  • scripts/lint-recursive-run.py
  • scripts/lint-recursive-run.ps1
  • scripts/recursive-review-bundle.py
  • scripts/recursive-review-bundle.ps1
  • scripts/recursive-closeout.py
  • scripts/recursive-closeout.ps1
  • scripts/recursive-router-init.py
  • scripts/recursive-router-init.ps1
  • scripts/recursive-router-probe.py
  • scripts/recursive-router-probe.ps1
  • scripts/recursive-router-configure.py
  • scripts/recursive-router-configure.ps1
  • scripts/recursive-router-invoke.py
  • scripts/recursive-router-invoke.ps1
  • scripts/recursive-router-resolve.py
  • scripts/recursive-router-resolve.ps1
  • scripts/recursive-router-validate.py
  • scripts/recursive-router-validate.ps1
  • scripts/recursive-subagent-action.py
  • scripts/recursive-subagent-action.ps1
  • scripts/recursive-lock.py
  • scripts/recursive-lock.ps1
  • scripts/verify-locks.py
  • scripts/verify-locks.ps1
  • scripts/check-reusable-repo-hygiene.py
  • scripts/check-reusable-repo-hygiene.ps1
  • .recursive/scripts/recursive-training-grpo.py
  • .recursive/scripts/recursive-training-grpo.ps1
  • .recursive/scripts/recursive-training-phase8-trigger.py
  • .recursive/scripts/recursive-training-phase8-trigger.ps1
  • .recursive/scripts/recursive-training-extract.py
  • .recursive/scripts/recursive-training-extract.ps1
  • .recursive/scripts/recursive-training-sync.py
  • .recursive/scripts/recursive-training-sync.ps1
  • .recursive/scripts/recursive-training-loader.py
  • .recursive/scripts/recursive-training-loader.ps1
  • .recursive/scripts/recursive-training-mcp.py
  • .recursive/scripts/recursive-training-mcp.ps1

Subskills

  • ../recursive-spec/SKILL.md
  • ../recursive-worktree/SKILL.md
  • ../recursive-debugging/SKILL.md
  • ../recursive-tdd/SKILL.md
  • ../recursive-review-bundle/SKILL.md
  • ../recursive-router/SKILL.md
  • ../recursive-subagent/SKILL.md
  • ../recursive-training/SKILL.md — extract durable experiential knowledge from completed runs

Use those subskills for their specialized discipline, but keep /.recursive/RECURSIVE.md as the single source of truth for the overall workflow contract.

Canonical routed CLI config lives under:

  • /.recursive/config/recursive-router.json
  • /.recursive/config/recursive-router-discovered.json

Router-aware delegated execution

When this skill or any recursive subskill is about to call another model or external CLI for delegated audit, review, bounded implementation, or other routed subagent work, re-read these routing files from disk immediately before choosing the CLI/model.

Do not hardcode provider or model strings for routed delegation when repo policy already defines the route or fallback behavior.

If routed delegation is in scope, prefer ../recursive-router/SKILL.md plus the canonical recursive-router-resolve and recursive-router-invoke scripts over bespoke model selection logic.

If routed delegation has been explicitly requested, or if recursive-router-resolve for the selected role returns external-cli, do not satisfy that delegated slot with local/self-audit work unless the effective route is fallback-local, local-only, blocked, or ask-user and that outcome is recorded. The controller may reject routed output after checking it against the actual files, diffs, and recursive artifacts, but unverified local work is not a substitute for a configured routed implementer, tester, reviewer, auditor, or planner role.

If a routed invocation returns success: false, exits nonzero, or reports issues that the bounded role is responsible for fixing, run an audit-repair-retry loop: preserve the failed attempt as evidence, tell the routed subagent exactly what to fix, rerun the same route after repair, then verify the new output locally before acceptance. Do not mark the delegated role or phase gate complete from a nonzero-exit routed attempt.

Before dispatching from an isolated worktree, make sure the routing policy and discovery inventory are present and current in that same worktree. Discovery inventory is often untracked, so a newly created worktree can have /.recursive/config/recursive-router.json without /.recursive/config/recursive-router-discovered.json; refresh discovery from the worktree or copy the current inventory from the controller/source repo before resolving or invoking a route, then record the paths and outcome in the phase artifact or action record.

Show full SKILL.md (700 more words)Show less

Optional add-on

  • recursive-benchmark stays outside the default recursive-mode package surface.
  • Install it only when a user explicitly asks to benchmark recursive-mode, preferably via find-skills or by adding its dedicated benchmark add-on package or repo source.

For audited phases, the installed workflow requires draft -> audit -> repair -> re-audit -> pass -> lock, with Audit: PASS required before Coverage/Approval may pass. Treat ## Worktree Diff Audit as phase-scoped: Phase 2 owns planning completeness plus expected product/worktree paths, Phase 3-4 own actual product/worktree drift, Phase 6 owns /.recursive/DECISIONS.md, Phase 7 owns /.recursive/STATE.md, and Phase 8 owns /.recursive/memory/**. New runs should use Workflow version: recursive-mode-audit-v2, where Phase 1 must include ## Source Requirement Inventory and Phase 2 must include ## Requirement Mapping, ## Plan Drift Check, and plan-stage ## Requirement Completion Status. For Phase 3, declare TDD Mode: strict|pragmatic. Strict mode requires concrete RED and GREEN evidence paths. Pragmatic mode requires an explicit exception rationale plus compensating evidence. For Phase 5, declare QA Execution Mode: human|agent-operated|hybrid. Do not fake human sign-off for agent-operated QA, and do not omit sign-off for human or hybrid QA. For delegated review, prefer .recursive/scripts/recursive-review-bundle.py or .recursive/scripts/recursive-review-bundle.ps1 so Phase 3.5 records a canonical Review Bundle Path. If meaningful subagent work contributes to a phase, record it under /.recursive/run/<run-id>/subagents/ and verify the action record against actual files, actual diffs, and actual recursive artifacts before accepting it. For review/audit delegation, prefer a stable reviewed artifact for Current Artifact rather than a mutable draft receipt. Audited phases must record Subagent Capability Probe, Delegation Decision Basis, ## Subagent Contribution Verification, and ## Requirement Completion Status. When subagents are available and the context bundle is complete, delegated audit/review is the default path. If the controller still uses Audit Execution Mode: self-audit, record a concrete Delegation Override Reason. Use the stronger requirement disposition fields in ## Requirement Completion Status: Changed Files, Implementation Evidence, Verification Evidence, Deferred By, Scope Decision, Blocking Evidence, and Addendum as appropriate to the chosen status. Do not mark implemented or verified without concrete Changed Files. Do not mark verified without distinct verification evidence. Treat addenda as authoritative effective inputs. When relevant addenda exist, list them under Inputs, re-read them in ## Effective Inputs Re-read, and reconcile them in ## Earlier Phase Reconciliation. Review bundles should include relevant addenda automatically, and the written Phase 3.5 review should cite bundle-grounded upstream artifacts, addenda, and changed files or code refs in the review narrative. 00-worktree.md is the source of truth for diff basis. Record baseline type/reference, comparison reference, normalized baseline/comparison, and normalized diff command; do not silently improvise a different basis later. Use Phase 8 to update skill memory under /.recursive/memory/skills/ when the run teaches the repo something durable about skill availability, fit, delegated-review quality, or skill-discovery outcomes. Phase 8 should also capture run-local skill usage before promoting any lesson into durable memory. If the run plans delegated review, review bundles, or smoke-harness portability work, read /.recursive/memory/skills/SKILLS.md and the relevant skill-memory shards before planning or auditing. If the run needs a specialized capability that is not already available, prefer the find-skills skill when it exists. Otherwise use the Skills CLI directly (npx skills find, npx skills add, npx skills check, npx skills update) and record the result when skill usage is relevant. Diff audit ignores incidental runtime byproducts such as __pycache__/, *.pyc, .pytest_cache/, .mypy_cache/, and .ruff_cache/ unless the repo intentionally tracks them. Treat Phase 6, Phase 7, and Phase 8 receipts as concise delta receipts that point to the final control-plane docs instead of restating large sections. Use .recursive/scripts/recursive-lock.py or .recursive/scripts/recursive-lock.ps1 as the normal locking path instead of manually editing Status, LockedAt, and LockHash. recursive-lock enforces monotonic phase gating: it will reject a lock attempt if any earlier phase that exists in the run directory is not already LOCKED. To reopen a locked artifact and reset it to DRAFT (which also invalidates downstream receipts), use the --reopen flag. On Windows, run Node/Vite/Vitest commands from the real worktree path rather than subst or mapped-drive aliases. When evaluating or building browser-local apps, consult references/local-first-web-app-checklist.md for common import/persistence/QA edge cases. If the repository being improved is itself a reusable skill/workflow repo, do not commit current-session run folders, evidence logs, review bundles, subagent action records, or temp-path references unless they are intentional fixtures; use scripts/check-reusable-repo-hygiene.py or scripts/check-reusable-repo-hygiene.ps1 before calling the repo clean.

© try-works, Apache-2.0. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

SKILL.md and 65 other files (scripts, references) in skills/recursive-mode of try-works/recursive-mode.

  • SKILL.md
  • agents/openai.yaml
  • references/agents-block.md
  • references/artifact-template.md
  • references/bootstrap/RECURSIVE.md
  • references/local-first-web-app-checklist.md
  • scripts/check-reusable-repo-hygiene.ps1
  • scripts/check-reusable-repo-hygiene.py
  • scripts/install-recursive-mode.ps1
  • scripts/install-recursive-mode.py
  • scripts/install-recursive-mode.sh
  • scripts/lint-recursive-run.ps1
  • scripts/lint-recursive-run.py
  • scripts/recursive-closeout.ps1
  • scripts/recursive-closeout.py
  • scripts/recursive-init.ps1
  • scripts/recursive-init.py
  • … and 49 more

Open the folder on GitHubat commit ff75bc7

Used in 1 other repository

We found 1 copy of this SKILL.md (exact, near-identical or edited) in other folders. This page covers the copy in try-works/recursive-mode, which our catalogue first saw on October 7, 2026.

Compare with similar skills

Recursive Mode 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.

Recursive Mode compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Recursive Mode this skilltry-works/recursive-mode136—~3.4kAutomated safety check: PassApache-2.0
Vercel Composition Patternssupabase/supabase111k59 repos~726Automated safety check: PassMIT
Finishing a Development Branchobra/superpowers296k5 repos~1.9kAutomated safety check: PassMIT
Typescript Advanced Typesrolling-scopes/rsschool-app10k25 repos~4.2kAutomated safety check: PassMPL-2.0
PR Babysitteropeninterpreter/openinterpreter69k3 repos~4.2kAutomated safety check: PassApache-2.0
Code Review ChecklistshareAI-lab/learn-claude-code78k5 repos~1.1kAutomated safety check: PassMIT

Similar skills

  • Vercel Composition Patterns

    supabase/supabase

    Official

    React composition patterns that scale. An agent skill from supabase/supabase.

    111k GitHub starsUsed in 59 repos~726 tokens
    DevelopmentAuto-check passed

  • Finishing a Development Branch

    obra/superpowers

    Walks the last step of a branch: confirm tests pass, detect the git environment, ask how to integrate, carry out your choice and clean up the worktree.

    296k GitHub starsUsed in 5 repos~1.9k tokens
    DevelopmentAuto-check passed

  • Typescript Advanced Types

    rolling-scopes/rsschool-app

    Master TypeScript's advanced type system including generics, conditional types, mapped types, template literals, and utility types for building type-safe applications.

    10k GitHub starsUsed in 25 repos~4.2k tokens
    DevelopmentAuto-check passed

  • PR Babysitter

    openinterpreter/openinterpreter

    Watches an open GitHub pull request until it merges, handling review comments, diagnosing CI failures and retrying flaky checks along the way.

    69k GitHub starsUsed in 3 repos~4.2k tokens
    DevelopmentAuto-check passed

  • Code Review Checklist

    shareAI-lab/learn-claude-code

    Reviews code against a five-part checklist covering security, correctness, performance, maintainability and testing, and reports findings in a fixed format.

    78k GitHub starsUsed in 5 repos~1.1k tokens
    DevelopmentAuto-check passed

  • Greploop

    onyx-dot-app/onyx

    Iteratively improves a PR (GitHub), MR (GitLab), or shelved changelist (Perforce) until Greptile gives it a 5/5 confidence score with zero unresolved comments.

    32k GitHub starsUsed in 4 repos~3.3k tokens
    DevelopmentAuto-check passed

More from try-works/recursive-mode

All 9 skills in this repo

  • Recursive Debugging

    try-works/recursive-mode

    A skill your agent uses when a recursive-mode requirement involves debugging a bug, test failure, or unexpected behavior.

    136 GitHub stars~3.4k tokensUpdated 2 mo ago
    Auto-check passed

  • Recursive Review Bundle

    try-works/recursive-mode

    A skill your agent uses when recursive-mode work needs a canonical delegated-review or audit handoff.

    136 GitHub stars~873 tokensUpdated 2 mo ago
    Auto-check passed

  • Recursive Subagent

    try-works/recursive-mode

    A skill your agent uses when recursive-mode work may benefit from delegated audit, review, or bounded implementation support.

    136 GitHub stars~3k tokensUpdated 2 mo ago
    Auto-check passed

  • Recursive TDD

    try-works/recursive-mode

    A skill your agent uses when implementing any code in recursive-mode Phase 3.

    136 GitHub stars~2.8k tokensUpdated 2 mo ago
    Auto-check passed

  • Recursive Worktree

    try-works/recursive-mode

    A skill your agent uses when starting any recursive-mode requirement to set up an isolated git worktree.

    136 GitHub stars~1.2k tokensUpdated 2 mo ago
    Auto-check passed

  • Recursive Spec

    try-works/recursive-mode

    Approval-gated, repo-aware requirements/spec authoring for new recursive-mode runs.

    136 GitHub stars~1.4k tokensUpdated 2 mo ago
    Auto-check passed

Categories

Questions about Recursive Mode

What does Recursive Mode do?

Repository workflow orchestration skill for staged implementation, locked artifacts, late-phase receipts, and durable memory maintenance. Recursive Mode is an agent skill from try-works/recursive-mode. Repository workflow orchestration skill for staged implementation, locked artifacts, late-phase receipts, and durable memory maintenance.

When should I use Recursive Mode?

Recursive Mode fits situations like: executing recursive-mode runs; locking a phase; verifying locks.

How do I install Recursive Mode in Claude Code?

Run `npx skills add try-works/recursive-mode --skill recursive-mode -a claude-code`. Or copy the skill folder (skills/recursive-mode in try-works/recursive-mode) into .claude/skills/recursive-mode in your project. Claude Code loads it when a task matches its description.

How do I install Recursive Mode in Codex?

Run `npx skills add try-works/recursive-mode --skill recursive-mode -a codex`. Or copy the skill folder (skills/recursive-mode in try-works/recursive-mode) into .agents/skills/recursive-mode in your project. Codex loads it when a task matches its description.

Can I use Recursive Mode 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 try-works/recursive-mode --skill recursive-mode -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/recursive-mode, .gemini/skills/recursive-mode, .github/skills/recursive-mode and .opencode/skills/recursive-mode in your project.

What does Recursive Mode need to run?

Going by SKILL.md and its folder, Recursive Mode needs PowerShell, Python and a shell for the scripts in its folder and the command-line tools its instructions call (npx, python, python3, bash and pwsh). Our summary lists: Python 3; Node.js; A Bash shell; PowerShell.

Does Recursive Mode access the network?

SKILL.md contains no URLs. Its commands use npx, which can reach the network depending on how they are called. This is read from the text; nothing was executed.

Is Recursive Mode 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. The check reads SKILL.md only: the scripts in the folder are not scanned, so read them before running anything.

What licence does Recursive Mode use?

Recursive Mode is published under the Apache-2.0 licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does Recursive Mode use?

About 3.4k tokens (SKILL.md is roughly 14k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full. Its references folder adds about 48k tokens, read only when the agent opens those files.

What are the alternatives to Recursive Mode?

Skills that share tags, products or a category with Recursive Mode: Vercel Composition Patterns (supabase/supabase, 111k stars), Finishing a Development Branch (obra/superpowers, 296k stars), Typescript Advanced Types (rolling-scopes/rsschool-app, 10k stars) and PR Babysitter (openinterpreter/openinterpreter, 69k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Recursive Mode?

try-works (a GitHub user) maintains it in try-works/recursive-mode, which has 136 GitHub stars. The repository holds 9 skills in this directory. The repository was last updated on July 24, 2026.

Source: try-works/recursive-mode on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.