validate
Validate a routine .md file's structure without calling the API.
Skill by hamzafer
Chalk Labs
Need help deploying validate?
Chalk Labs sets up agents like this inside real teams: connected to your tools, tested on your data, with a human in the loop where it matters.
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/hamzafer-validate
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
Use cases: backend, agent-building
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
Network hosts it talks to 1
codeload.github.com· found in install command
Files it touches
~/.claude/skills/validate
What's inside
- Skill name
validate- Description
- Validate a routine .md file's structure without calling the API. Checks frontmatter, cron schedule rules, model field, allowed_tools, sources, mcp_connections, and prompt body. Invoke when the user says "validate <file>", "lint <file>", or "validate" (no file = validate everything in the workspace).
- License
- MIT
- Instructions
- 77 lines in SKILL.md · view on GitHub
Files (1)
- SKILL.md
SKILL.md (first lines)
# claude-routines:validate
Lint a routine file before deploying. Never calls the API. Run this whenever you're about to `deploy`.
## Per-file checks
Run these in order. Stop the file at the first error; collect all warnings.
### 1. Frontmatter parses
Must be valid YAML between the opening and closing `---`. If parsing fails, report the YAML error and stop.
### 2. Required fields
- `name` (string, non-empty)
- `env_id` (string starting with `env_`)
- Exactly one of `cron` (string) or `run_once_at` (RFC3339 UTC timestamp). Both present is an error (`conflicting fields`). Neither present is an error (`missing trigger`).
### 3. Cron rules
When `cron` is set:
- Standard 5-field cron in UTC.
- Minimum interval is 1 hour. The minute field must be a single literal value (e.g. `0` or `30`). Reject patterns like `*/30 * * * *` or comma-lists like `0,30 * * * *` (under 1h gap).
- Hour field may use lists or ranges; that's fine since gaps are at least 1h.
### 4. run_once_at rules
Must be a future RFC3339 UTC timestamp like `2027-01-01T00:00:00Z`. Past timestamps are a warning (the API still accepts them; they fire immediately).
### 5. enabled
Must be a bool if present. Default `true`.
### 6. model
If set, must be one of: `claude-opus-4-7`, `claude-opus-4-7[1m]`, `claude-sonnet-4-6`, `claude-haiku-4-5`. Anything else is an error.
### 7. allowed_tools
Must be a list of strings. Warn if empty (inherits nothing). Warn loudly if it contains every default write-tool (`Bash`, `Write`, `Edit`, `NotebookEdit` all present): that's the silent-expansion default-set; suggest an explicit trim.
### 8. sources
If present, must be a list of objects with `url` (string starting `https://github.com/`) and optional `allow_unrestricted_git_push` (bool).
### 9. mcp_connections
If present, must be a list of objects with `connector_uuid` (lowercase v4 UUID), `name`, `url`, and optional `permitted_tools`.
### 10. Snippet includes (optional feature)
If the prompt body contains `{{include <path>}}` lines: the file at `<path>` must exist relative to the routine file's directory or the workspace root. Included files must NOT themselves contain `{{include}}` directives (no nesting).
### 11. Prompt body
Must be non-empty after stripping whitespace.
## Output format
```
OK personal/morning-brain-digest.md
ERR personal/oslo-apartment-hunter.md
- error: cron "*/30 * * * *" violates minimum 1-hour interval
- warning: allowed_tools includes Bash, Write, Edit, NotebookEdit (full default-set). Consider trimming.
```
When called without a file argument, validate every `.md` file under common routine locations (`routines/`, `personal/`, or anywhere with frontmatter). Skip READMEs. Print a one-line summary per file.
Final line: `N files OK | M files with errors | W warnings`.
## See also
- `dry-run deploy <file>` to preview the API payload after validation passes
- `deploy <file>` to actually shipFrom hamzafer/claude-routines, MIT.
More from hamzafer/claude-routines
- 0.540score
run
hamzafer
Fire a Claude Code routine immediately, bypassing its schedule.
SkillCFreshness: C★ 10Claude Code +9 - 0.540score
pull
hamzafer
Fetch all Claude Code routines from the cloud and write each one to a local .md file.
SkillCFreshness: C★ 10Claude Code +9 - 0.540score
update
hamzafer
Update an existing Claude Code routine via the read-modify-write protocol.
SkillCFreshness: C★ 10Claude Code +9 - 0.540score
bulk
hamzafer
Apply an operation across many routine .md files at once.
SkillCFreshness: C★ 10Claude Code +9 - 0.540score
orphans
hamzafer
Find local routine .md files whose trigger_id no longer exists in the cloud (probably deleted via the web UI).
SkillCFreshness: C★ 10Claude Code +9 - 0.540score
deploy
hamzafer
Deploy a Claude Code routine from a local .md file.
SkillCFreshness: C★ 10Claude Code +9 - 0.540score
diff
hamzafer
Compare a local routine .md file against its cloud state field by field.
SkillCFreshness: C★ 10Claude Code +9 - 0.540score
dry-run
hamzafer
Preview the exact API body that "deploy <file>" would send, without calling the API.
SkillCFreshness: C★ 10Claude Code +9 - 0.540score
get
hamzafer
Fetch a single Claude Code routine by its trigger_id and write it to a local .md file.
SkillCFreshness: C★ 10Claude Code +9 - 0.540score
list
hamzafer
List all Claude Code routines for the current user.
SkillCFreshness: C★ 10Claude Code +9 - 0.540score
create
hamzafer
Create a new Claude Code routine from a local .md file.
SkillCFreshness: C★ 10Claude Code +9 - 0.535score
delete
hamzafer
Handle deletion requests for Claude Code routines.
SkillCFreshness: C★ 10Claude Code +9
Signals
- Stars
- 10
- Forks
- 0
- Open issues
- 0
- Last commit (this folder)
- May 14, 2026 (5 months ago)
- Last release
- Apr 27, 2026 (6 months ago)
From GitHub (fetched Oct 24, 2026).
Details
- License
- MIT
- Repository
- github.com/hamzafer/claude-routines
- First seen
- Oct 24, 2026
- Last checked
- Oct 24, 2026
- Source
- GitHubgithub-graphql
More from hamzafer
- 0.540score
run
hamzafer
Fire a Claude Code routine immediately, bypassing its schedule.
SkillCFreshness: C★ 10Claude Code +9 - 0.540score
pull
hamzafer
Fetch all Claude Code routines from the cloud and write each one to a local .md file.
SkillCFreshness: C★ 10Claude Code +9 - 0.540score
update
hamzafer
Update an existing Claude Code routine via the read-modify-write protocol.
SkillCFreshness: C★ 10Claude Code +9 - 0.540score
bulk
hamzafer
Apply an operation across many routine .md files at once.
SkillCFreshness: C★ 10Claude Code +9 - 0.540score
orphans
hamzafer
Find local routine .md files whose trigger_id no longer exists in the cloud (probably deleted via the web UI).
SkillCFreshness: C★ 10Claude Code +9 - 0.540score
deploy
hamzafer
Deploy a Claude Code routine from a local .md file.
SkillCFreshness: C★ 10Claude Code +9
Done for you
Rather not set it up yourself?
Chalk Labs scopes, builds and runs agents for your team: the task audit, the eval harness, the integrations and the weekly ops report. Fixed scope, priced before we start.
See how we build AI agents →Read the case studies
Chalk Labs is a two-founder AI and growth studio. We build AI agents, CRMs and MVPs for Web3 and AI companies, and have shipped 50+ product and token launches.
- launches shipped
- 50+
- launches shipped
- media placements
- 500+
- media placements
- founders on every account
- 2
- founders on every account
Press we've placed: Forbes · CoinDesk · Decrypt · Cointelegraph · The Block
Rahil Jain, Co-founder, Chalk Labs. Ex-Goldman Sachs engineer. You talk to the person who builds it.
Get validate running in your business
Tell us what you want to automate. A founder replies within one working day with a plan, not a pitch deck.
Cite this page
validate skill. Agent Mart. https://agentmart.dev/skills/hamzafer-validate (last updated 2026-10-24)