Tier-2 Review: internal-comms (Slug: internal-comms)

What We Attempted

We evaluated the internal-comms skill, which claims to help Heads of People Ops, BizOps leads, or Internal Communications owners draft and sequence internal change-management communications (e.g., re-org announcements, tool rollouts, policy changes, layoffs, acquisition closes). The skill promises to combine Prosci ADKAR and Kotter’s 8-step change models with deterministic Python tools to produce a sequenced touchpoint calendar, a primary announcement, an audience-segmented FAQ, and manager cascade talking points. We attempted to install the skill and run a smoke-invocation test to verify basic functionality.

What Failed

Both tests in our harness failed:

• Install (fail): The SKILL.md does not document any install command. The skill is described as a Python script using only stdlib dependencies, but it is not packaged as a pip-installable module or any other standard distribution format. There is no requirements.txt, no setup.py, and no explicit instruction for how to obtain or install the script. The test harness could not proceed with installation.

• Smoke-invocation (partial): The SKILL.md describes a script that uses only stdlib, which suggests invocation might work if the script exists. However, no explicit CLI entry point or usage example is provided. The documentation assumes the script is named internal_comms.py but does not confirm this. Without a clear command or path, the harness could not execute the script. One test passed partially because the description mentions stdlib-only dependencies, implying no external packages are needed, but this is insufficient for a reliable smoke test.

Key blocker: No documented install command or CLI entry point in SKILL.md.

What We Observed

The skill has a well-structured prompt and clear triggers (e.g., "all-hands announcement", "town-hall script", "change comms", "internal newsletter", "re-org announcement"). The composite score of 4.5/5.0 reflects strong trigger clarity (5.0), output specificity (4.5), scope precision (4.5), self-containment (4.5), and reusability (4.0). The auto-summary notes "excellent skill with clear triggers, specific outputs, and precise scope." However, these ratings are based on the skill's design and documentation, not on actual execution. The skill appears to be a Python script with stdlib-only dependencies, which is theoretically portable and self-contained, but the lack of installation instructions and CLI entry point makes it unusable in practice without additional effort from the user.

Theoretical vs. Practical Rating

The rating of 4.5/5.0 is theoretical until a physical re-run resolves the failures. The skill’s concept is sound—combining ADKAR and Kotter models with deterministic tools for internal comms is a valuable niche. However, the current implementation is incomplete from a deployment perspective. Users would need to guess how to obtain the script, name it, and invoke it. This undermines the "reusability" dimension (currently rated 4.0) and the "self-containment" dimension (4.5), as the skill cannot be used without significant manual setup.

Is the Skill Still Valuable in Principle?

Yes, the skill remains valuable in principle. The combination of Prosci ADKAR and Kotter’s 8-step model is a proven framework for change management, and automating the production of sequenced touchpoint calendars, announcements, FAQs, and manager talking points would save significant time for internal comms professionals. The industry-tuning via --profile (tech-startup, scaleup, enterprise, public sector) adds further utility. If the installation and invocation issues are resolved—by adding a simple install command, a CLI entry point, and a usage example—this skill could be highly effective. The core logic and prompt design are strong; the gap is purely in packaging and documentation.

Final note: This is honest signal, not a takedown. The skill’s idea is promising, but it needs a concrete delivery mechanism to realize its potential. Until then, the high composite score should be taken with caution.

Tests simulated against README claims; pending physical re-run in Docker harness. Ran 2026-05-31.

Overall: broken. 0 tests passed, 1 partial, 1 failed; key blocker: no documented install command or CLI entry point in SKILL.md.

Inferred dependencies: python>=3.10, stdlib-only.

Test	Status	Notes
install	fail	No install command documented in SKILL.md; skill is a Python script with stdlib-only dependencies, not a pip package.
smoke-invocation	partial	SKILL.md describes a script that uses only stdlib, so invocation may work if script exists, but no explicit CLI entry point or usage example is provided; assumes script is named internal_comms.py.