ai-physea-liminality declares 27 capabilities across 27 tools, with 2 critical issues.
78/100 exposureHigh exposure
Manifest retrieved 2026-09-14 from tools/list on https://liminality.physea.ai/mcp, published in the official MCP registry as ai.physea/liminality 1.0.5. Not verified by a maintainer.
Issues
These describe what this server's manifest declares. A product can provide controls that MCP has no
way to express — token revocation, an admin console, audit logs — and this page cannot see them. The absence of a
declaration is not evidence that a control is absent.
critical · no-revocation · All 27 tools
27 of 27 tools declare no way for the customer to switch them off once granted. This applies to every tool because MCP has no field in which to declare a revocation mechanism; the product may provide one outside MCP that a manifest cannot show.
critical · irreversible-uncapped · 1 of 27 tools
1 tool is classified as irreversible and declares no limit.
high · no-expiry · All 27 tools
27 of 27 tools declare no expiry. This applies to every tool because MCP has no field in which to declare an expiry; the product may provide one outside MCP that a manifest cannot show.
high · no-audit-trail · All 27 tools
27 of 27 tools declare no record of their actions that the customer can read. This applies to every tool because MCP has no field in which to declare an audit requirement; the product may provide one outside MCP that a manifest cannot show.
high · uncapped-write · 19 of 27 tools
19 tools can modify data and declare no limit.
low · uncapped-read · 7 of 27 tools
7 tools can read data and declare no limit.
Declared capabilities
How each tool was read. declared means the server's own annotation said so — self-reported and
unverified. inferred is from the tool's name. assumed means nothing indicated either way.
Tool
Read as
Basis
solve
write
assumed
get_capabilities
read
inferred
get_settings
read
inferred
set_settings
write
inferred
get_result_detail
read
inferred
export_result
write
assumed
compress_context
write
assumed
delegation
write
assumed
handoff
write
assumed
ask_form
write
inferred
apply_form
write
assumed
report_outcome
write
assumed
report_feedback
write
assumed
apply_grade
write
assumed
optimize
write
assumed
algorithms
write
assumed
task
read
inferred
recover
read
inferred
runs
delete
inferred
research
write
assumed
composio_connect
write
inferred
register_asset
write
assumed
set_priorities
write
inferred
taxis
read
inferred
catalog
read
inferred
setup
write
assumed
manage
write
assumed
Answering this question about your own agent? Media Yard LLC runs a
fixed-scope Agent Permission Audit: the same reading applied to your agent surface, delivered as a dated
artifact you can hand to a customer's security review. See what it includes and costs, or write to
support@leashkit.com.
Buying one does not change this page, its score or its place in the index — see Independence.
Maintain this server? If this page misreads your manifest, or your server has changed,
submit its current tools/list response and it will be re-scored and dated.
Corrections are published alongside the original, not in place of it.