Accept 100M/100K token notation for templatable token-limit fields

[Copilot]

max-effective-tokens and max-daily-effective-tokens were still validated as generic templatable_integer, so values like 100M/100K failed schema validation (as seen in the compile failure). This change aligns those fields with the token-limit schema that already supports K/M suffix notation, including lowercase variants.

• Schema alignment

  • Rewired top-level schema fields:
    • max-effective-tokens → #/$defs/max_effective_tokens_limit
    • max-daily-effective-tokens → #/$defs/max_daily_effective_tokens_limit
  • This enables string token notation (K/M, case-insensitive) and preserves expression support.

• Parser schema tests

  • Converted prior suffix rejection expectations into acceptance expectations.
  • Added explicit mixed-case coverage across both fields (100M, 100k, 100K, 100m patterns).
  • Updated zero/negative expectation cases to match these defs (0 invalid; -1 sentinel behavior where defined).

• Example

  max-effective-tokens: 100K
  max-daily-effective-tokens: 100m

[pelikhan]

@copilot refactor and reuse same def

[Copilot]

Refactored to reuse a shared schema def (positive_effective_tokens_limit) across both token-limit fields in commit 4f94739.

[Copilot]

Pull request overview

Aligns the main workflow frontmatter schema so max-effective-tokens and max-daily-effective-tokens accept the same token-limit notation already supported elsewhere (e.g., 100K/100M, case-insensitive) while preserving GitHub Actions expression support.

Changes:

• Rewired top-level schema properties to reference token-limit-specific $defs instead of templatable_integer.
• Refactored/expanded $defs for effective-token limits to clearly separate “positive limit” vs “disabled via negative / -1”.
• Updated parser schema validation tests to expect suffix strings and -1 sentinel values to pass, and 0 to fail.

Show a summary per file

File	Description
pkg/parser/schemas/main_workflow_schema.json	Repoints max-effective-tokens / max-daily-effective-tokens to token-limit defs and introduces a shared “positive effective tokens limit” helper to allow K/M suffix + expressions.
pkg/parser/schema_test.go	Updates schema validation tests for 0, -1, and K/M-suffixed string acceptance (including lowercase support).
.github/skills/agentic-workflows/SKILL.md	Adds additional referenced docs to the skill’s “load these files” list.

Copilot's findings

Tip

Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

• Files reviewed: 3/3 changed files
• Comments generated: 3