john/claude-engineering-plugin
1
0
Fork 0
Code Issues Pull Requests Actions 3 Packages Projects Releases Wiki Activity

fix(create-agent-skills): remove literal dynamic context directives that break skill loading

Browse Source 
The `create-agent-skills` SKILL.md contained literal `!`command`` dynamic
context injection directives as documentation examples. Claude Code's
preprocessor executes these directives as plain text before the skill
content is sent to the model — it does not parse markdown, so fenced code
blocks and inline code spans offer no protection.

When loading this skill, the preprocessor attempted to run `command`,
`gh pr diff`, and `gh pr diff --name-only` as shell commands, producing:

    Bash command failed for pattern "!`command`":
    (eval):1: redirection with no command

This caused the skill to fail silently or error on every invocation.

Fix: replace the literal directives with a prose description of the
syntax, add a warning about the preprocessor behavior, and link to
`references/official-spec.md` § "Dynamic Context Injection" for the
concrete example (reference files are loaded on-demand by Claude via the
Read tool and are not preprocessed).

Upstream context:
- anthropics/claude-code#27149 (closed, wontfix): preprocessor correctly
  executes bare text; workaround is to describe the syntax in prose
- anthropics/claude-code#28024 (closed as duplicate of #13655)

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
This commit is contained in:
Vahe Kirakosyan
2026-03-11 11:36:46 +01:00
parent 42cc74c7c7
commit 4b4d1ae270
1 changed files with 3 additions and 14 deletions
Download Patch File Download Diff File Expand all files Collapse all files

17
plugins/compound-engineering/skills/create-agent-skills/SKILL.md
View File

@@ -97,22 +97,11 @@ Access individual args: `$ARGUMENTS[0]` or shorthand `$0`, `$1`, `$2`.
### Dynamic Context Injection
The `` !`command` `` syntax runs shell commands before content is sent to Claude:
Skills support dynamic context injection: prefix a backtick-wrapped shell command with an exclamation mark, and the preprocessor executes it at load time, replacing the directive with stdout. Write an exclamation mark immediately before the opening backtick of the command you want executed (for example, to inject the current git branch, write the exclamation mark followed by `git branch --show-current` wrapped in backticks).
```yaml
---
name: pr-summary
description: Summarize changes in a pull request
context: fork
agent: Explore
---
**Important:** The preprocessor scans the entire SKILL.md as plain text — it does not parse markdown. Directives inside fenced code blocks or inline code spans are still executed. If a skill documents this syntax with literal examples, the preprocessor will attempt to run them, causing load failures. To safely document this feature, describe it in prose (as done here) or place examples in a reference file, which is loaded on-demand by Claude and not preprocessed.
## Context
- PR diff: !`gh pr diff`
- Changed files: !`gh pr diff --name-only`
Summarize this pull request...
```
For a concrete example of dynamic context injection in a skill, see [official-spec.md](references/official-spec.md) § "Dynamic Context Injection".
### Running in a Subagent